Citation
The epizootiology and pathogenicity of Haemoproteus meleagridis Levine, 1961, from Florida turkeys

Material Information

Title:
The epizootiology and pathogenicity of Haemoproteus meleagridis Levine, 1961, from Florida turkeys
Creator:
Atkinson, Carter Tait, 1954-
Publication Date:
Language:
English
Physical Description:
xix, 294 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Dissertations, Academic -- Veterinary Medicine -- UF ( mesh )
Parasitic Diseases -- pathology ( mesh )
Parasitic Diseases, Animal ( mesh )
Turkeys -- parasitology -- Florida ( mesh )
Veterinary Medicine ( mesh )
Veterinary Medicine thesis Ph.D ( mesh )
City of Gainesville ( local )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1985.
Bibliography:
Bibliography: leaves 276-293.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Carter Tait Atkinson.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
022763766 ( ALEPH )
16936047 ( OCLC )
ACW5301 ( NOTIS )

Downloads

This item has the following downloads:

EJ0IDPM7Z_JXOFE6.xml

AA00011797_00001.pdf

AA00011797_00001_0071.txt

AA00011797_00001_0075.txt

AA00011797_00001_0013.txt

AA00011797_00001_0082.txt

AA00011797_00001_0297.txt

AA00011797_00001_0024.txt

AA00011797_00001_0203.txt

AA00011797_00001_0043.txt

AA00011797_00001_0079.txt

AA00011797_00001_0258.txt

AA00011797_00001_0255.txt

AA00011797_00001_0070.txt

AA00011797_00001_0277.txt

EJ0IDPM7Z_JXOFE6_xml.txt

AA00011797_00001_0140.txt

AA00011797_00001_0267.txt

AA00011797_00001_0170.txt

AA00011797_00001_0059.txt

AA00011797_00001_0279.txt

AA00011797_00001_0227.txt

AA00011797_00001_0048.txt

AA00011797_00001_0304.txt

AA00011797_00001_0110.txt

AA00011797_00001_0293.txt

AA00011797_00001_0190.txt

AA00011797_00001_0126.txt

AA00011797_00001_0315.txt

AA00011797_00001_0033.txt

AA00011797_00001_0144.txt

AA00011797_00001_0095.txt

AA00011797_00001_0316.txt

AA00011797_00001_0261.txt

AA00011797_00001_0225.txt

AA00011797_00001_0049.txt

AA00011797_00001_0215.txt

AA00011797_00001_0302.txt

AA00011797_00001_0002.txt

AA00011797_00001_0103.txt

AA00011797_00001_0092.txt

AA00011797_00001_0283.txt

AA00011797_00001_0186.txt

AA00011797_00001_0230.txt

AA00011797_00001_0012.txt

AA00011797_00001_0078.txt

AA00011797_00001_0054.txt

AA00011797_00001_0161.txt

AA00011797_00001_0301.txt

AA00011797_00001_0173.txt

AA00011797_00001_0235.txt

AA00011797_00001_0123.txt

AA00011797_00001_0250.txt

AA00011797_00001_0216.txt

AA00011797_00001_0084.txt

AA00011797_00001_0154.txt

AA00011797_00001_0056.txt

AA00011797_00001_0146.txt

AA00011797_00001_0196.txt

AA00011797_00001_0088.txt

AA00011797_00001_0180.txt

AA00011797_00001_0099.txt

AA00011797_00001_0252.txt

AA00011797_00001_0211.txt

AA00011797_00001_0257.txt

AA00011797_00001_0055.txt

AA00011797_00001_0138.txt

AA00011797_00001_0087.txt

AA00011797_00001_0167.txt

AA00011797_00001_0281.txt

AA00011797_00001_0209.txt

AA00011797_00001_0194.txt

AA00011797_00001_0076.txt

AA00011797_00001_0222.txt

AA00011797_00001_0001.txt

AA00011797_00001_0234.txt

AA00011797_00001_0274.txt

AA00011797_00001_0061.txt

AA00011797_00001_0224.txt

AA00011797_00001_0057.txt

AA00011797_00001_0303.txt

AA00011797_00001_0111.txt

AA00011797_00001_0247.txt

AA00011797_00001_0273.txt

AA00011797_00001_0286.txt

AA00011797_00001_0010.txt

AA00011797_00001_0165.txt

AA00011797_00001_0053.txt

AA00011797_00001_0081.txt

AA00011797_00001_0027.txt

AA00011797_00001_0127.txt

AA00011797_00001_0020.txt

AA00011797_00001_0240.txt

AA00011797_00001_0145.txt

AA00011797_00001_0028.txt

AA00011797_00001_0205.txt

AA00011797_00001_0223.txt

AA00011797_00001_0207.txt

AA00011797_00001_0130.txt

AA00011797_00001_0046.txt

AA00011797_00001_0191.txt

AA00011797_00001_0064.txt

AA00011797_00001_0176.txt

AA00011797_00001_0278.txt

AA00011797_00001_0218.txt

AA00011797_00001_0296.txt

AA00011797_00001_0219.txt

AA00011797_00001_0120.txt

AA00011797_00001_0246.txt

AA00011797_00001_0253.txt

AA00011797_00001_0025.txt

AA00011797_00001_0312.txt

AA00011797_00001_0298.txt

AA00011797_00001_0245.txt

AA00011797_00001_0244.txt

AA00011797_00001_0311.txt

AA00011797_00001_0200.txt

AA00011797_00001_0182.txt

AA00011797_00001_0086.txt

AA00011797_00001_0137.txt

AA00011797_00001_0091.txt

AA00011797_00001_0011.txt

AA00011797_00001_0116.txt

AA00011797_00001_0269.txt

AA00011797_00001_0105.txt

AA00011797_00001_0115.txt

AA00011797_00001_0294.txt

AA00011797_00001_0231.txt

AA00011797_00001_0199.txt

AA00011797_00001_0213.txt

AA00011797_00001_0295.txt

AA00011797_00001_0221.txt

AA00011797_00001_0090.txt

AA00011797_00001_0100.txt

AA00011797_00001_0052.txt

AA00011797_00001_0287.txt

AA00011797_00001_0271.txt

AA00011797_00001_0300.txt

AA00011797_00001_0112.txt

AA00011797_00001_0264.txt

AA00011797_00001_0233.txt

AA00011797_00001_0030.txt

AA00011797_00001_0131.txt

AA00011797_00001_0051.txt

AA00011797_00001_0275.txt

AA00011797_00001_0183.txt

AA00011797_00001_0210.txt

AA00011797_00001_0289.txt

AA00011797_00001_0026.txt

AA00011797_00001_0181.txt

AA00011797_00001_0021.txt

AA00011797_00001_0155.txt

AA00011797_00001_0109.txt

AA00011797_00001_0083.txt

AA00011797_00001_0039.txt

AA00011797_00001_0148.txt

AA00011797_00001_0018.txt

AA00011797_00001_0239.txt

AA00011797_00001_0009.txt

AA00011797_00001_0162.txt

AA00011797_00001_0093.txt

AA00011797_00001_0147.txt

AA00011797_00001_0016.txt

AA00011797_00001_0306.txt

AA00011797_00001_0232.txt

AA00011797_00001_0214.txt

AA00011797_00001_0134.txt

AA00011797_00001_0058.txt

AA00011797_00001_0141.txt

AA00011797_00001_0133.txt

AA00011797_00001_0077.txt

AA00011797_00001_0124.txt

AA00011797_00001_0172.txt

AA00011797_00001_0308.txt

AA00011797_00001_0195.txt

AA00011797_00001_pdf.txt

AA00011797_00001_0151.txt

AA00011797_00001_0156.txt

AA00011797_00001_0307.txt

AA00011797_00001_0285.txt

AA00011797_00001_0158.txt

AA00011797_00001_0040.txt

AA00011797_00001_0029.txt

AA00011797_00001_0132.txt

AA00011797_00001_0185.txt

AA00011797_00001_0168.txt

AA00011797_00001_0202.txt

AA00011797_00001_0107.txt

AA00011797_00001_0114.txt

AA00011797_00001_0201.txt

AA00011797_00001_0096.txt

AA00011797_00001_0037.txt

AA00011797_00001_0276.txt

AA00011797_00001_0280.txt

AA00011797_00001_0212.txt

AA00011797_00001_0122.txt

AA00011797_00001_0050.txt

AA00011797_00001_0188.txt

AA00011797_00001_0270.txt

AA00011797_00001_0229.txt

AA00011797_00001_0266.txt

AA00011797_00001_0288.txt

AA00011797_00001_0142.txt

AA00011797_00001_0313.txt

AA00011797_00001_0068.txt

AA00011797_00001_0268.txt

AA00011797_00001_0062.txt

AA00011797_00001_0164.txt

AA00011797_00001_0177.txt

AA00011797_00001_0060.txt

AA00011797_00001_0113.txt

AA00011797_00001_0163.txt

AA00011797_00001_0065.txt

AA00011797_00001_0314.txt

AA00011797_00001_0305.txt

AA00011797_00001_0309.txt

AA00011797_00001_0017.txt

AA00011797_00001_0260.txt

AA00011797_00001_0290.txt

AA00011797_00001_0204.txt

AA00011797_00001_0106.txt

AA00011797_00001_0042.txt

AA00011797_00001_0263.txt

AA00011797_00001_0031.txt

AA00011797_00001_0259.txt

AA00011797_00001_0045.txt

AA00011797_00001_0152.txt

AA00011797_00001_0206.txt

AA00011797_00001_0006.txt

AA00011797_00001_0129.txt

AA00011797_00001_0102.txt

AA00011797_00001_0128.txt

AA00011797_00001_0217.txt

AA00011797_00001_0063.txt

AA00011797_00001_0159.txt

AA00011797_00001_0166.txt

AA00011797_00001_0041.txt

AA00011797_00001_0098.txt

AA00011797_00001_0236.txt

AA00011797_00001_0117.txt

AA00011797_00001_0150.txt

AA00011797_00001_0256.txt

AA00011797_00001_0008.txt

AA00011797_00001_0189.txt

AA00011797_00001_0003.txt

AA00011797_00001_0220.txt

AA00011797_00001_0291.txt

AA00011797_00001_0241.txt

AA00011797_00001_0243.txt

AA00011797_00001_0292.txt

AA00011797_00001_0184.txt

AA00011797_00001_0237.txt

AA00011797_00001_0171.txt

AA00011797_00001_0038.txt

AA00011797_00001_0121.txt

AA00011797_00001_0238.txt

AA00011797_00001_0108.txt

AA00011797_00001_0015.txt

AA00011797_00001_0197.txt

AA00011797_00001_0074.txt

AA00011797_00001_0022.txt

AA00011797_00001_0248.txt

AA00011797_00001_0125.txt

AA00011797_00001_0032.txt

AA00011797_00001_0047.txt

AA00011797_00001_0035.txt

AA00011797_00001_0310.txt

AA00011797_00001_0228.txt

AA00011797_00001_0044.txt

AA00011797_00001_0139.txt

AA00011797_00001_0187.txt

AA00011797_00001_0208.txt

AA00011797_00001_0135.txt

AA00011797_00001_0198.txt

AA00011797_00001_0101.txt

AA00011797_00001_0226.txt

AA00011797_00001_0072.txt

AA00011797_00001_0242.txt

AA00011797_00001_0085.txt

AA00011797_00001_0136.txt

AA00011797_00001_0014.txt

AA00011797_00001_0080.txt

AA00011797_00001_0149.txt

AA00011797_00001_0193.txt

AA00011797_00001_0249.txt

AA00011797_00001_0251.txt

AA00011797_00001_0073.txt

AA00011797_00001_0174.txt

AA00011797_00001_0118.txt

AA00011797_00001_0160.txt

AA00011797_00001_0005.txt

AA00011797_00001_0034.txt

AA00011797_00001_0004.txt

AA00011797_00001_0104.txt

AA00011797_00001_0284.txt

AA00011797_00001_0067.txt

AA00011797_00001_0119.txt

AA00011797_00001_0143.txt

AA00011797_00001_0178.txt

AA00011797_00001_0179.txt

AA00011797_00001_0262.txt

AA00011797_00001_0094.txt

AA00011797_00001_0097.txt

AA00011797_00001_0254.txt

AA00011797_00001_0036.txt

AA00011797_00001_0157.txt

AA00011797_00001_0265.txt

AA00011797_00001_0066.txt

AA00011797_00001_0272.txt

AA00011797_00001_0175.txt

AA00011797_00001_0069.txt

AA00011797_00001_0153.txt

AA00011797_00001_0019.txt

AA00011797_00001_0023.txt

AA00011797_00001_0192.txt

AA00011797_00001_0169.txt

AA00011797_00001_0282.txt

AA00011797_00001_0299.txt

AA00011797_00001_0089.txt

AA00011797_00001_0007.txt


Full Text
52
Throughout the study at Fisheating Creek, average
monthly temperatures never fell below 60o F.
Transmission and Vector Abundance
Cu 1 i coi de s eden i C. h i nman i C. arboricol a and C.
knowltoni were the only species captured in sufficient
numbers in Bennett traps, to be implicated as potential
vectors at Paynes Prairie and Fisheating Creek. Because
representatives of all 4 species had distinctive peaks
in activity at sunset, biting activity was plotted for
each as the logjo 0f the number captured during the peak
Bennett trap run for a sampling night, plus 1.
Paynes Prairie. During the 2-year study at Paynes
Prairie, biting collections and light trap collections for
each species were positively correlated (Figures 13, 14).
Specimens of edeni were active at variable levels
throughout the year. Bennett trap and light trap
catches were usually lowest during the cooler, winter
months between January and March (Figures 13, 14, 18).
Transmission of 1C me 1eagrid i s to sentinel birds began soon
after the biting activity and abundance of (C eden i
increased in late April, 1983 and 1984, with the onset
of warmer spring weather. Peaks in the transmission of H.
me 1 eagridis to sentinel turkeys corresponded to minor peaks


56
Between April, 1983, and May, 1984, unengorged
specimens of Cu I icoi des captured at Paynes Prairie and
Fisheating Creek were pooled and inoculated into
domestic turkey poults. At Paynes Prairie, 2 of 9 pools of
C. eden i totaling 343 individuals, were positive for H.
me 1eagrid i s resulting in an estimated minimum yearly
prevalence of 0.58% (Table 8). The positive pools were
collected in November, 1983, and May, 1984, during periods
of active, natural transmission of the parasite.
At Fisheating Creek, 17 of 49 pools of C^ edeni,
totaling 816 individuals, were positive for me 1eagridis,
resulting in an estimated minimum yearly prevalence of
2.08% (Table 9). Positive pools were collected in
April, July, September and November, 1983, and February,
March and May, 1984, during active natural transmission of
the parasite.
Unengorged individuals of other species of Cu 1 icoi des
were not captured in sufficient numbers to make regular
attempts at isolation. Pools of O hinmani, C. arboricola
and C^ knowltoni, collected at unequal intervals throughout
the year were negative for me 1eagridis (Table 10).


Figure Page
36 Average parasitemias £or high dose birds, low
dose birds and control birds 130
37 Average weights o£ high dose birds, low dose
birds and control birds 132
38 Average tarsometatarsal lengths of high dose
birds, low dose birds and control birds 134
39 Average hematocrits for high dose birds, low
dose birds and control birds 136
40 Average plasma protein concentrations for
high dose birds, low dose birds and control
birds 138
41 Average hemoglobin concentrations for high
dose birds, low dose birds and control birds .. 140
42 Three-day-old schizont 142
43 Five-day-old schizont 142
44 Five-day-old schizont 142
45 Eight-day-old megaloschizont 142
46 Eight-day-old mega 1oschizont 142
47 Fourteen-day-old mega 1oschizont 142
48 Fourteen-day-old mega 1oschizont 144
49 Fourteen-day-old mega 1oschizont 144
50 Seventeen-day-old mega 1oschizont 144
51 Seventeen-day-old megaloschizont 146
52 Seventeen-day-old mega 1oschizont 146
53 Disrupted muscle fiber from pectoral muscle
of a turkey with a 3-day-old experimental
infection of Haemoproteus me 1eagridis 148
x i i i


63
Table 8. Yearly prevalence of Haemoproteus me 1 eaer i dis
in specimens of Cu 1 ic o id e s e d e ni at Paynes
Prairie.
Month
Year
Poo 1 s
# Flies
Isolations
Apr i 1
1983
1
35
0
June
1983
1
83
0
July
1983
1
19
0
August
1983
1
10
0
November
1983
1
20
1
March
1984
1
12
0
April
1984
1
12
0
May
1984
2
152
1
Total :
9
343
2
Minimum YearIy Prevalence:
0.58%


273
Jaker, J.R. 1963. The host restriction o£ Haemoproteus
columbae. J. Protozool. 15: 334-335.
Bao, L. 1959. A cytological study o£ the early oocysts of
seven species of PI a smod i um and the occurrence o£
post-zygotic meiosisC Parasito 1ogy. 49: 559-535.
Barnard, D.R. and R.H. Jones. 1980. Die! and seasonal
patterns of flight activity of ceratopogonids in
northeastern Colorado. Environ. Entomol. 9: 446-451.
Bates, M. 1949. The natural history of mosquitoes. The
Macmillan Co. New York. 379 pp.
Becker, E.R., W.F. Hollander and W.H. Patti 1 1o. 1956.
Naturally occurring P1 a smodium and Haemop r oteus
infection in the common pigeon. J. Pa r a sito 1. T2 :
474-478.
Bennett, G.F. 1960. On some ornithophilic
blood-sucking Dptera in Algonquin Park, Ontario,
Canada. Can. J. Zool. 38: 377-389.
Bennett, G.F., and A.G. Campbell. 1972. Avian
Haemoproteidae. I. Description of Haemoproteus fa 11 isi
n.sp. and a review of the haemoprote i as ol the tamily
Turdidae. Can. J. Zool. 50: 1269-1275.
Bennett, G.F., and A.G. Campbell. 1975. Avian
Leucocytozo idae. 1. Morphometric variation in three
species of L eucocytozoon and some taxonomic
implications. Can. Zool. 53: 800-312.
Bennett, G.F., and R.F. Coombs. 1975. Ornithophilic
vectors or avian hematozoa in insular
Newfoundland. Can. J. Zool. 53: 1241-1246.
Bennett, G.F., and A.M. Falls. 1960. Blood parasites of
birds of Algonquin Park, Canada, and a discussion of
their transmission. Can. J. Zool. 38: 261-273.
Bennett, G.F., P.C.C. Garnham and A.M. Fall is. 1965. On
the status of the genera Leucocy tozoon Ziemann 1898
and Ha emo proteus Kruse 1890 ( Haemo s po ridida:
Leucocytozoidae and Haemoprote i dae) Can. J. Zool.
43: 927-932.


5
diverse than was realized previously. There is no reason
to suggest that h a emo p r o t e i d s are any less so.
Accordingly, several authors have proposed that this genus
should be divided into 2 genera (Bennett et al., 1965).
They suggested that species transmitted by hippoboscid
flies, with large oocysts containing several hundred blunt
sporozoites, would remain in the genus Haemop r ot eu s
Those species transmitted by ceratopogonids with small
oocysts containing fewer than 100 pointed sporozoites,
would be placed in the genus Parahaemoproteus.
The size and presence or absence of septa in the
exoerythrocytic schizonts has been proposed as another
criterion for separating the 2 genera (Garnham, 1966).
One form of schizont, found in birds infected with H.
co1umbae, H. 1ophor t yx H. pa 1umbis and f ringi 11ae,
is sausage or oval shaped, lacks cytomeres, and occurs
in endothelial cells in a variety of tissues including
lung, liver, spleen, kidney, bone marrow, cecum and heart.
The other type, described in birds infected with H,
ga r nhami, is large and focal with diameters of 200 urn
or larger and contains numerous cytomeres, separated from
one another by septa (Garnham, 1966). Unfortunately,
schizont morphology has not correlated perfectly with
vector or oocyst size. This is exemplified by the
exoerythrocytic stages of fU f r i ngi 11ae (Khan and Falls,
1969). This species, which develops in Cu 1 icoi des has


184


24
calculated when more than 1 sample was taken during a
sentinel period (Bid1ingmeyer, 1969).
Between December, 1982, and November, 1984, collecting
trips were made approximately once a month to Lykes
Fisheating Creek Wildlife Management Area at Palmdale,
Florida, 310 km SSE of Paynes Prairie. This area has
one of the most dense Wild Turkey populations in the state
(Powell, 1967; L.E. Williams, pers comm.). Earlier
sentinel work by Forrester (unpublished) had shown that
H. meleagridis was transmitted year-round in this area.
A study site was selected in a live oak harrmock,
surrounded by cypress, at the edge of a creek swamp, 5
km SSE of Palmdale. During each collecting trip, a Bennett
trap was operated in the middle level of the canopy for
1-3 consecutive evenings and occasionally at night and
during the day. The New Jersey light trap, supplemented
with dry ice, was operated 1 night/trip in the middle
level of the canopy at a second hoist, 30 meters from
the first. Three to 5, 2-week-old domestic poults were
also exposed at the study site for the duration of each
trip. They were housed, as described earlier, I meter
above the ground in a sentinel cage. The birds were
transported to and from Gainesville in a screened,
vector-proof cage and bled as described earlier to diagnose
infections.


60
Table 5. Engorged specimens of Cu 1 i coi des captured in
Bennett traps at Fisheating Creek, December 1982 -
November 1984. Traps were operated on 47
different evenings for a total of 108 hours.
Spec ies
Tota 1
% Total
C. edeni
2,038
79.6
C. hinmani
475
18.6
C. knowltoni
35
1.4
C. arboricola
12
0.5
C. baueri
1
0.04
2,561 100.0
Total


239
amorphous, eosinophilic material that resembled the
material observed in necrotic mega 1 os ch i zonts of H.
me 1eagridis.
The mega 1oschizonts described by Nair and Forrester
(unpublished) in skeletal muscle of a naturally infected
Wild Turkey are identical in structure and size to
meg a I oschizonts of me 1eagrid i s from experimentally
infected turkeys. Many of the cysts observed by Nair
and Forrester contained disorganized masses of
basophilic material. Since the tissue had been frozen for
several weeks prior to fixation, degenerative changes in
the megalosch i zont morphology may be an artifact of the
manner in which the tissue had been handled. The discovery
of these forms in a naturally infected Wild Turkey
indicates that the site of development and the
associated host response are not artifacts related to
the way the experimental turkeys were infected.
Mi 1tgen et al. (1931) noted the similarity between the
exoerythrocytic stages of desseri and Arthrocyst i s
ga 1 1 i an organism of uncertain taxonomic status described
by Levine et al. (1970) in chickens from India, and
suggested that the 2 organisms may be synonymous. Other
descriptions of cyst-forming organisms have been
reported from a variety of avian hosts in several families
of birds. Most infections have been characterized by
the presence of large, intramuscular schizonts, similar in


236
Since 14-day-old meg a 1 oschizonts contained numerous
discrete bodies that resembled the cytomeres described
in developing megaloschizonts of Leucocytozoon, the
terminology as defined by 3ray ( 1960) has been applied
to megaloschizonts of me 1eagridis.
The severe lameness and extensive myopathy that
occurred in infected birds during development of the small,
first-generation schizonts of me 1eagridis is unusual for
haemospor i d i an infections. The early, pre-erythrocytic
stages of most species of PI a smodium do not elicit any host
reaction and later exoetythrocytic schizonts appear to
be pathogenic only during unusual circumstances when the
parasite is exceptionally infectious to the host (Huff,
1969). Little host reaction has been observed in
association with the small endothelial schizonts of H.
f ringi 1 Iae H. coIumbae H. pa 1umbis H lophortyx, H.
ne 11 i on i s and macea 1 1 um i (Khan and Falls, 1969;
Moharrmed, 1965; Ahmed and Mohammed, 1977; 3aker, 1966a;
O'Roke, 1930; Sibley and Werner, 1984; Greiner, 1971).
Garnham (1966) reported cellular infiltration similar to
acute interstitial pneumonia in the alveolar septa of Rock
Doves with heavy, early infections of co1umbae, but did
not provide any details of the infection. Of the 3 genera
of avian haemosporidians the exoerythrocytic stages of
Leucocytozoon have been documented as the most
pathogenic (Lund and Farr, 1965). However, most pathology


51
September, 1983, and below average during July and August,
1983, and July, 1984 (Figure 16).
Two peaks in transmission occurred in 1983 at each
site. A small peak was evident between July and August and
again between November and December at Site A. Peaks at
Site B occurred 1 to 2 months earlier between May and June
and between October and November. In 1984, transmission at
both sites began in mid-April and continued until the
end of the study in July.
Fisheating Creek. Between February, 1983, and
November, 1984, 52 of 66 (78.8%) sentinel turkeys
exposed during collecting trips to Fisheating Creek became
infected with me 1eag r i d i s Exposures as short as 24
hours in November and December, 1983, and January, March
and June, 1984, were sufficient to infect 100% of the 3 or
4 sentinel turkeys that were taken during each trip.
Between June, 1983, and September, 1984, 50 100% of
the sentinel birds exposed during each trip became infected
with me 1eag rid i s (Figure 18). Transmission did not
occur in February, March and April, 1983, during periods of
abnormally cool and wet weather (Figures 19, 20).
Transmission was not detected in November, 1984, when
monthly precipitation was above average and mean
temperatures were slightly below normal (Figures 19, 20).


253
membranes and a thickened, ostniophilic inner membrane
composed of 2 unit membranes in close apposition to one
another (Bradbury and Roberts, 1970; Sterling, 1972;
Sterling and Aikawa, 1973). By contrast, the inner
membrane of mature gametocytes of me 1eagrid i s consisted
of a single unit membrane that was thicker and more
osmiophilic than the outer 2 membranes. It is unlikely that
the difference was a result of poor fixation, since
other membranous structures within the gametocytes and
their host cells were well preserved.
Sterling and Aikawa (1973) suggested that the
double inner membrane of gametocytes of co1umbae was
a remnant of the intramembranous complex of merozoites.
3ased on observations of PIasmod i urn spp., they suggested
that this 2-layered complex was retained in merozoites that
had recently invaded a red blood cell if the merozoite was
to become a gametocyte. De differentiation of the
complex occurred if asexual, schizogonic stages developed.
The similarities in structure and development among most
haemosporidian parasites make it unlikely that the
thickened, inner pellicle of me 1eagridis had a different
origin. Perhaps the 2-layered intramembranous complex
of merozoites of me 1 e a g t i d i s fused to form the
single, thickened layer that ivas observed in gametocytes.
Other studies of macrogametocytes of Haemoproteus and
Leucocytozoon have noted the presence of amorphous,


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EJ0IDPM7Z_JXOFE6 INGEST_TIME 2012-09-24T12:48:00Z PACKAGE AA00011797_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


Figure 26. An intact meg a 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection .
The megaloschizont is surrounded by
dark-staining areas of calcification. Parallel
lines of dark-staining granules are present in
1 muscle fiber adjacent to the schizont
(arrow). Hematoxylin and eosin. Bar = SO um.
Figure 27. A serial section of the megaloschizont
illustrated in Figure 26. The section is
stained with von Kossa's stain for calcium.
There is a close correspondence between the
dark-sta ining deposits in Figure 26 and the
dark-sta ining calcium deposits in Figure
27. Bar = 50 um.


248
cage compartments to one another, the inevitable fecal
contamination that occurred in the food and water and
successful isolation of Salmone11 a from representatives of
each experimental group suggests that all the birds were
infected. Since birds vary in their output of
Sa 1 mone 1 la organisms from day to day, the low number of
Salmone11 a isolations from cloacal swabs of each group at 4
and at 3 weeks post-infection is not surprising (Williams,
1978). It is significant that the only birds to develop
clinical salmonellosis were those in the high dose group.
They exhibited signs of infection between 12 and 28 days
post-infection when stress from the lb me 1eag r i dis
infection reached its peak. Perhaps the Haemoproteus
infection weakened the high dose group sufficiently to make
t h em susceptible to Salmon ella and other secondary
bacterial and fungal infections as well. S a 1mo n e 1 la i s
frequently isolated from commercial feeds that use
animal products to boost protein levels (Williams, 1978).
It is likely that the turkeys used in the experiment
acquired their infections from the unmedicated game bird
chow they were fed.
Most authors have considered Haemoproteus to be a
relatively benign parasite (3ennett et al., 1982; Kemp,
1978; Fallis and Desser, 1977; Levine, 1961). Considering
how prevalent Haemoproteus is in many bird populations
(Bennett, 1982), the few isolated reports of pathogenic


151
Host Spec ificity
Parasitemia
Haemoproteus me 1eagr idis was successfully transmitted
to 1 of 2 Ring-necked Pheasants, 1 of 2 Chuckars and 2
turkeys that acted as positive controls in the first series
of experimental infections. One of the 2 inoculated
Guineafowl was found dead 6 days post-infection (DPI). One
day prior to death, the bird appeared healthy. At
necropsy, all organs appeared normal. Histological
sections of liver, lung, heart, kidney, gizzard, pancreas,
duodenum, cecum and brain were unremarkable. Skeletal
muscle and spleen were not fixed. The surviving inoculated
Guineafowl and all negative controls failed to develop
patent infections throughout the duration of the
exper ¡merit.
All infected birds became patent at 17 DPI. The 2
infected turkeys reached an average peak parasitemia of 813
parasites per 10,000 red blood cells at 22 DPI. The
parasitemia rapidly fell by 28 DPI to 33 parasites per
10,000 red blood cells and remained at levels less than 15
parasites per 10,000 red blood cells for the duration of
the experiment (Figure 59). By contrast, the positive
Chuckar reached a lower peak parasitemia of 162
parasites per 10,000 red blood cells at 17 DPI. A
rapid drop to 23 parasites per 10,000 red blood cells


40
cleared in acetone and embedded in Spurt 1s resin.
Ultrathin sections were cut on glass knives, stained with
5% (w/v) aqueous uranyl acetate and 2% (w/v) Reynold's
lead citrate and examined with a Hitachi HU-1 IE electron
microscope.
Mature gametocytes were fixed by drawing blood from
the wing vein of a turkey infected with H_^ me 1eagr i d i s
into a syringe containing the primary fixative. The
resulting clots were diced in the primary fixative, fixed
for 1 hour at room temperature, washed with 3, 10-minute
changes of buffer, post-fixed with osmium for 1 hour at
room temperature, washed with 3 more 10-minute changes
of buffer and dehydrated and embedded as described above.
Several drops of fresh blood from the same turkey were
placed on a glass slide in a humidity chamber to allow
the gametocytes to exf1 age 11 ate. Two minutes and 3 minutes
after the drops were made, the clots were flooded with
primary fixative and processed as above.
Specimens of edeni that had engorged on a turkey
with a heavy H^ me 1eag ridis infection, 3 and 6 days
earlier, were dissected in a drop of Aed e s aegypt i
Ringer's. The midguts were carefully removed, flooded
with a drop of primary fixative and processed through
the first series of washes. To facilitate handling and
to prevent loss during subsequent steps, the midguts were
embedded in warm 2% agar made with 0.1 M sodium cacodylate


260
of the gametocyte in sheets and whorls and floats free
in the host cell cytoplasm. Sterling (1972) did not
observe the detachment of the outer layer from gametocytes
of metc'nnikovi but did note that it was missing once
the gametocytes became extracellular. Desser (1972a)
described mu 1ti I aminar, membranous structures external
to maturing, intracellular gametocytes of ve 1ans, but
failed to associate their appearance with the loss of
the outer layer of the pellicle in extracellular
gametocytes. He suggested instead that they were
altered bands of microtubules from host red blood cells
that normally aid in maintaining the shape of the cell.
Micrographs published by Bradbury and Roberts (1970) and
Aikawa and Sterling (1974), as well as observations from
this study, leave little doubt that these membranous sheets
and whorls originated from the outer layer of the
gametocyte pellicle. This process appears to be limited to
the genus Haemoproteus and has not been reported from
studies of PI a smodium o r Leucocytozoon (Aikawa and
Sterling, 1974a). Ribosome-1ike granules similar to those
observed on some membranous whorls in this study were also
observed by 3radbury and Roberts (1970) on whorls that
detached from gametocytes of ib co1umbae. The significance
of the granules and their origin are unknown.


208


29
containing 10% turkey serum in a glass tissue grinder
in wet ice for several minutes. The resulting slurry
was then inoculated as described into uninfected poults.
Estimates of the total number of sporozoites inoculated
were made with a hemocytometer. Permanent sporozoite
preparations were made by mixing the slurry 1:10 with
turkey serum. A drop was smeared on a glass slide, air
dried and fixed and stained as above. Measurements of
stained sporozoites were made as above.
Estimation of Prevalence
During operation of the Bennett trap, only 50-70%
of captured specimens of Cu 1 icoi des normally took a blood
meal from the bait turkey. Unengorged specimens of
Cu 1 icoi des were identified by wing pattern, grouped by
species and ground in lots of 10-20 in Aede s aegyp ti
ringers or RPMI tissue culture media, as described
earlier. The slurry from each pool was inoculated
intraperitonea 11 y into separate 1- to 2-week old turkey
poults. These birds were bled as described earlier to
diagnose any infections. Minimum yearly prevalence of
H. me 1eag rid i s in pools of naturally infected eden i
was calculated for Paynes Prairie and Fisheating Creek
as the total number of positive pools / total number of
specimens of edeni.


285
Gonder, R. 1915. On the transmission of Haemoproteus
col umbae. Rep. Dir. Vet. Res. Pretoria 3 & 4: 627-032.
Greiner, E.C. 1971. The comparative life histories of
Haemoproteus sacharov i and Haemoproteus macca 11umi in
the Mo u r ning Dove (Zenaida macroura) FF.D.
Thesis. University of Nebraska, Lincoln. TT5 pp.
Greiner, E.C. 1975. Prevalence and potential vectors of
Haemoproteus in Nebraska mourning doves. J. Wildl.
Dis.1l: 150-156.
Greiner, E.C., E.S. Eveleigh and W.M. Boone. 1978.
Ornithophi I ic Cu 1 ico i des spp. (Dptera:
Ce ratopogon i dae ) from New Brunswick, Canada, and
implications of their involvement in haemoproteid
transmission. J. Med. Ent. 14: 701-704.
Greiner, E.C., and D.J. Forrester. 1980. Haemop r o t eu s
me 1eag r i d i s Levine 1961: redescription and
d e v e1opmen ta 1 morphology of the gametocytes in
turkeys. J. Parasitol. 66: 652-658.
Greiner, E.C., and A.A. Kocan. 1977. Leucocytozoon
(Haemospori da; Leucocytozoidae) of the Falcon¡formes.
Can. J. Zool. SS: 761-770.
Haberkorn, A. 1968 Zur hormonellen Bee inf1ussung von
Haemoproteus-Infectionen. Z. Parasitenk. 31:
108-112.
Hanson, H.C., N.D. Levine, C.W. Kossack, S. Kantor and
L.J. Stannard. 1957. Parasites of the mourning dove
(Zenaida macroura carolinensis) in Illinois. J.
Parasitol. 43: 186-193.
Hayes, R.O. 1953. Determination of a physiological saline
solution for Aedes aegypti. (L. ) J. Econ. Ent. 46:
624-627.
Herman, C.M. 1938. The epidemiology of malaria in
eastern redwings (Agelaius p. phoeniceus). Am. J.
Hyg. 28: 232-243.
Herman, C.M., and G.F. Bennett. 1976. Use of sentinel ducks
in ep i zooti o 1ogica1 studies of anatid blood protozoa.
Can. J. Zool. 54: 1038-1043.


91
DEC JAN 1 EEB 1 MAR' APR MAY JUN 1 JUL 1 AUG 1 SEP 1 OCT 1 NOV 1 DEC 1 JAN 1 FEB MAR 1 APR 1 MAY JUN 1 JUl AUG 1 SEP 1 OCT 1 NOV 1 DEC 1 JA <
2
82 1 1983
11984
rM
Culicoides hinmani
o 1
/
LOG
xJJ
v
y .a\
0
#1cf'
.
V
DEC 1 JAN 1 FEB 1 MAR 1 APR1 MAT 1 JUN 1 JUL AUG 1 SEP 1 OCT 1 NOV 1 DEC JAN 1 EEB 1 MAP 1 APR 1 MAY 1 JUN 1 JU. 1 AUG 1 SEP 1 OCT 1 NOV 1 DEC 1 JAN


274
had villous processes extending £rom their external
surfaces. They also noted that the megaloschizonts were
divided into thick-walled compartments resembling tubes.
Megaloschizonts of me 1 e ap,r i d i s differ from those
observed by Desser (1970a) and Gardiner et at. (1984). The
megaloschizonts observed in this study were
extracellular and had thick, laminated walls composed of
electron-dense, granular material. Material examined by
Gardiner et al. (1984) was fixed initially in buffered
formalin and then post-fixed with osrni um-d i chromate and
processed for electron microscopy. The poor
preservation of cellular detail in their samples make
comparisons with the megaloschizonts of me 1e agridis
difficult. However, the villous processes on the cyst
walls and cornpa r trnen ta 1 i za t i on of megaloschizonts by the
thick wall were never observed in megaloschizonts of H.
me 1eagridis .
Desser (1970a) described merozoite formation in
megaloschizonts of simo n di by fragmentation of the
cytomere cytoplasm. By contrast, Gardiner et al. (1984)
reported that merozoite development in megaloschizonts from
Northern Bobwhites occurred by budding into an interior
vacuole within the cytomere cytoplasm rather than by the
protrusion of developing merozoites from the cytomere
surface. Neither process was observed in
megaloschizonts of me 1eagridis.


WEEK
LOG|Q (Parasitemia + I)
o ro oj
\L I


107
Plasma protein concentration. Statistical analysis of
plasma protein concentration revealed that all variables in
the model statement were significant (p < .0125).
When comparisons were made by week, all 3 groups were
significantly different at I week PI. Control birds had
the highest average plasma protein concentration and
high dose birds had the lowest. 3y 2 weeks PI, average
plasma protein concentrations were significantly greater
for high dose birds than either low dose or control birds.
The 3 experimental groups were not significantly different
at 3 and 4 weeks PI. At 5 and 6 weeks PI, high dose birds
had average plasma protein concentrations that were
significantly higher than low dose and control birds. The
3 groups were not significantly different at 7 and 8 weeks
PI (Figure 40).
When comparisons were made within groups, the low dose
birds had significantly lower average plasma protein
concentrations at weeks 0, 1 and 3 PI than they did at
weeks 2, 4, 5, 6, 7 and 8 PI.
Average plasma protein concentrations in the high dose
group were significantly lower at 1 and 3 weeks PI than
they were at 0, 4, 7 and 8 weeks PI. Average
concentrations at the latter 4 weeks were significantly
lower than those at 2 and 5 weeks PI. Average values at 0
and 3 weeks PI were also significantly lower than those at
4, 5, 6, 7 and 8 weeks PI.


I certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Philosophy.
Emeritus Professor of Veterinary
Med icine
I certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Philosophy.
Mart in D
/ /
4a
Young /7"
Professor
Research
Medicine
of Veterinary
This dissertation was submitted to the Graduate Faculty of
the College of Medicine and was accepted as partial
fulfillment of the requirements for the degree of Doctor of
Philosophy.
De c emb e r, 1985
Dean, Graduate School


14
Desser (1972a, 1972b) suggested that the large
crystalloid inclusions in ookinetes of ve 1ans originated
from amorphous, dense lipid inclusions in the
macrogametocyte. He felt that this material was converted
to a crystalline form by addition of a protein component
by a network of endoplasmic reticulum that surrounded
the precursor material. In contrast, crystalloid particles
in ookinetes of co1umbae first appeared between the
lamellae of the endoplasmic reticulum (Gallucci, 1974b).
Other differences between the ookinetes of these 2 species
occur in the fine structure of their anterior ends.
Gallucci (1974b) found a distinctive conoid in ookinetes
of co1umbae. Ookinetes of other haemosporidia ,
including fH ve 1ans, seem to lack this structure, although
observations among some species are limited (Gallucci,
1974b; Desser, 1972b). Other anterior organelles such
as conoidal rings, the apical pore, the canopy and
subpe 11 i cu 1ar microtubules appear to be present in both
H. co1umbae and ve 1ans, although Gallucci (1974b) and
Desser (1972b) interpret their micrographs differently.
Differences in interpretation are particularly evident
in descriptions of the polar ring and ribs in ookinetes
of these 2 species. Desser (1972b) interpreted structures
resembling the ribs described by Gallucci (1974b) in H.
co1umbae as an empty subpe 11 i cu 1 ar space and stated that
the subpe11 icu1 ar microtubules in ve 1 an s arose from


Figure 30. A spleen section from a high dose bird that
died spontaneously at 22 days pos t-i n £ ec t i on .
Numerous masses of dark pigment are
scattered throughout the tissue.
Hematoxylin and eos in. Bar = 20 urn.
Figure 31. A schizont in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection. The host cell has a
pycnotic nucleus (arrow). The schizont is much
smaller than megaloschizonts from skeletal
muscle in the same bird and lacks a thick,
hyaline wall. Hematoxylin and eosin. Bar = 10
urn.
Figure 32. Schizonts in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection. One schizont contains elongate
zoites (large arrow). Free merozoites
(small arrow) are scattered throughout the
tissue. Hematoxylin and eosin. Bar = 10 urn.


Figure 11. Modified New Jersey suction trap catches of
specimens of Cu 1 icoi de s ed en i at Fisheating
Creek during tlie Ma r ch, April and May, 1983
collecting trips. ( = canopy trap;0"'0 =
ground trap). Rising and setting suns indicate
the 2-hour sampling period that included
dawn or dusk.


UNIVERSITY OF FLOI^
!| !l I4 6413


242
Pathology
Experimental infections of Haemoproteus meleagridis
produced a moderate to severe myositis and myocarditis
in domestic turkeys. The inflammatory reactions were
associated with the development of first and
second-generation schizonts. The lameness exhibited by
infected birds and the gross and microscopic lesions
were similar to naturally acquired infections of
Arthrocystis gall i (Levine et a I ., 1970; Opitz et a 1 ,
1982). Both Levine et al. (1970) and Opitz et al. (1982)
noted extensive muscle necrosis, inflammation and
hemorrhage around the mega 1oschizonts. Opitz et al. ( 1982)
also observed dystrophic calcification of necrotic
muscle fibers and the formation of scar tissue. Neither
Opitz et al. ( 1 982 ) or Levine et al. (1970) reported
the thrombus formation or cicculatory disturbances that
were observed in this study. It is likely that they
developed in association with the acute and chronic
infla mm atory responses to intact and ruptured
mega 1 osch i zonts and the associated necrotic changes in
surrounding muscle fibers.
The distribution and morphology of second-generation
megalosch i zonts observed in the pathogenicity experiment
was identical to mega 1 oschizonts found during earlier
experimental infections. In addition, small,
thin-walled schizonts that contained both elongate


206


30
Pathogen icity
Experiment 1 Pathology
Thirty-six, 1-day-old, female, broad-breasted white
turkey poults were obtained in October, 1984. They were
housed together in a brooder in a vector-proof room for
7 days, then banded with metal wing tags and randomly
assigned to 3 experimental groups. Birds in the first
group were inoculated IP with separate pools of 5 C edeni
that had taken blood meals from 4 domestic poults infected
with hC me 1eag ridis The specimens of Cu 1 icoi des were
captured in October, 1984, at Fisheating Creek with a
Bennett trap, held for 8-9 days at room temperature to
allow development of sporozoites and then ground by hand
in a glass tissue grinder for several minutes in an ice
bath. The insects were triturated in RPMI tissue culture
medium containing 10% turkey serum. The 4 domestic poults
used to infect the wild-captured specimens of eden i
had acquired their infections from a previous exposure
at Fisheating Creek. Capture of specimens of C^ edeni
occurred during 2 evenings on days 9 and 10 of patency
when most gametocytes were fully mature. All 4 birds
had similar parasitemias. Engorged C^ eden i from both
trap nights were assigned randomly to pools used to infect
the experimental poults. Sporozoites from I pool of
specimens of C^ eden i were counted with a hemocy tometer


196
77


283
Eve, J.H., F.E. Kellog and F.A. Hayes. 1972. Blood
parasitisms of wild turkeys in the southeastern
U.S.A. J. Am. Vet. Med. Assoc. 161: 638-640.
Falls, A.M., and G.F. Bennett. I960. Description of
Haemoproteus canachites n. sp. (Sporozoa:
Haemoproteidae) and sporogony in Culi coi des (Dptera:
Ceratopogonidae). Can. J. Zool. 38: 455-464.
Fa 1 1 is, A.M., and G.F. Bennett. 1961. Sporogony of
Leucocytozoon and Haemoproteus in simuliids and
ceratopogonids anda revised classification of
the Haemosporidiida. Can. J. Zool. 39: 215-228.
Fallis, A.M., and G.F. Bennett. 1966. On the epizootiology
of infections caused by Leucocytozoon si mo n di in
Algonquin Park, Canada. Can~ 7~. Zoo 1 T": 101-112.
Fallis, A.M., and S.S. Desser. 1977. On species of
Leucocytozoon Haemop t o t e u s and Hepa t ocy s t i s I n
"Parasitic Frotozoa" vol. 3. J.P. K r eie r, ed.
Academic Press, New York. pp. 239-266.
Fallis, A.M., and D.M. Wood. 1957. Biting midges
(Dptera: Ceratopogon i dae) as intermediate hosts for
Haeinoproteus of ducks. Can. J. Zool. 35: 425-435.
Farmer, J.N. 1965. Gizzard lesions associated with
Ha emo proteus sacharovi infections of pigeons.
Proc. Iowa Acad. Sci. 71: 537-542.
Forrester, D.J., E.C. Greiner and M.K. Kigaye. 1977. Avian
Haemoproteidae. 7. A review of the haemoprot e i ds of
the family Ciconiidae (storks) and descriptions of
Ha emop roteus brodkorbi sp.nov. and Haeinoproteus
pe ircei sp.nov. Can. J. Zool. 55: 1268-127TI
Forrester, D.J., L.T. Hon, L.E. Williams Jr. and
D.H. Austin. 1974. Blood protozoa of wild
turkeys in Florida. J. Protozool. 21: 494-497.
Frenkel, J.K. 1973. Toxoplasmosis: parasite life cycle,
pathology, and immunology, in "The Coccidia", D.M.
Hammond and P.L. Long, eds. University Park Press,
Baltimore. 482 pp.
Frenkel, J.K., and J.P. Dubey. 1975. Hammond i a hamnondi
gen.nov., sp.nov., from domestic cats, a new coccidian
related to Toxoplasma and Sarcocystis. Z. Parasitenk.
46: 3-12.


To Mom and Dad


Figure 15. Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Paynes Prairie. Normals are based on 30
year averages for each month between 1951
and 1980 (Climatological Data: Florida,
1982, 1983, 1984).


10
9
8
7
6
5
4
3
2
I
0
Culicoides orboricolo


X = 19.2
s

8 *
-140 -120 -100 -80 -60 ^40 -20 0 20 40 60 80 100 120 140
CAPTURE TIME


264
Bradbury and Trager (1968b) described the polarization
of mi crogametocytes of H^ col umbae into 2 halves 1
that contained organelles such as axonemes, mitochondria
and food vacuoles and another that contained remnants of
the microgametocyte nucleus. They observed the
disintegration of the nuclear membrane and a
condensation of nuclear material around the bases of
developing axonemes. The condensed masses of nuclear
material were subsequently surrounded by a membrane and
incorporated into microgametes that "peeled" from the
mi ergametocy te.
Desser (1972a) described the formation of atypical
cent r¡oles and electron dense nuclear plaques in
mi c r ogame t ocy t e s of ve 1 an s He did not observe the
polarization of exflagellating microgametocytes
described by Bradbury and Trager (1968b), but did describe
the breakdown of the nuclear membrane and the subsequent
condensation of masses of chromatin around the bases of
developing axonemes.
Observations from this study are very limited, but
they are most similar to the process described by Aikawa
and Sterling (1974a). The nuclear membrane remained intact
in m i c r og ame t oc y t e s of me 1 eag r i d i s throughout the
process of exflagellation, but was often difficult to
discern. During the final budding of microgametes, the
nucleus appeared to be stretched to their base. The


167
Table 21. Average adjusted measurements of host cells
infected with microgametocytes. Each variable
was divided by the average value of the
same variable from uninfected cells of the same
spec i es.
Variable Chuckar Pheasant Turkey
Ce 11 Length
1.17
(0.09)+
1.03
(0.07)
1.11
(0.06)
Cell Width
1 .02
(0.08)
1.17
(0.08)
I 1 1
(0.09)
Ce11 Area
1.18
(0.10)
1.21
(0.08)
1.27
(0.11)
Nucleus Length
0.92
(0.11)
0.82
(0.11)
0.99
(0.11)
Nucleus Width
0.84
(0.11)
0.98
(0.17)
0.95
(0.10)
Nucleus Area
0.77
(0.17)
0.31
(0.15)
0.94
(0.16)
NDR*
0.89
(0.15)
0.28
(0.21)
0.42
(0.26)
N =
15
15
15
+ Standard Deviation
* Nuclear Displacement Ratio (I = no lateral displacement)
(0 = lateral displacement to cell margin)


Figure 89. A cytomere with budding merozoites (arrows). A
nucleus (N) in the process o£ division is
constricted into 2 lobes. Budding
merozoites contain a large, electron-lucent
vacuole (V), rhoptries (R) and a mitochondrion
(M). Merozoites have a pellicle composed of an
outer unit membrane and a discontinuous
inner layer consisting of 2 unit membranes
in close apposition to one another (small
arrows). X 58,000.
Figure 90. Mature merozoite. The merozoite has a
large, e1ectron-1ucent vacuole (V), paired
rhoptries (R) and 3 anterior polar rings
(arrows). X 58,000.
Figure 91. Mature merozoite. The merozoite has a
large, e 1 ectron-1ucent vacuole (V), paired
rhoptries (R) and 3 anterior polar rings
(arrows). X 58,000.


Pae
Figure
24 A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection 120
25 A degenerating megaloschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infect ion 120
26 An intact mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection.
Hematoxylin and eosin stain 122
27 A serial section of the mega 1oschizont
illustrated in Figure 26. von Kossa's stain
for calcium 122
28 A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection 124
29 A venule blocked by a thrombus in pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection 124
30 A spleen section from a high dose bird that
died spontaneously at 22 days post-infect ion .. 126
31 A schizont in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection 126
32 Schizonts in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection 126
33 Nodular infiltrate from pectoral muscle of a
high dose bird that was killed at 8 weeks
post-infection 128
34 A thrombus, composed of fibrinous material,
adjacent to the remnants of a degenerating
megaloschizont 128
35 A mass of degenerating muscle fibers,
infiltrated with macrophages and heterophils .. 128
x i i


lbUUUIdlUIdLitildl
59
Table 4. New Jersey light trap collections Fisheating
Creek, December 1982 November 1984. Total
catch was made on 28 different nights of
t rapping.
:c i es
Total
% Total
i ns ignis
7,940
44.5
eden i
7,279
40.8
knowltoni
1,356
7.6
s t e11ifer
742
4.2
arborico1 a
287
1.6
hinmani
129
0.7
pusi1 I us
60
0.3
ousairani
28
0.2
baueri
15
0. 1
paraensis
14
0. 1
deb i 1 i pa 1 pus
4
0.02
haematopotus
2
0.01
b i ck1eyi
1
0.01
17,857
100.00
Total


35
Engorged specimens of ed e n i and hinmani were
collected and held as described earlier for 7 to 9 days.
They were randomly assigned to pools of 48 specimens of
C. eden i and 33 specimens of h i nman i triturated as
described earlier and inoculated IP into 6, 5-day-old,
female broad-breasted white turkey poults. One pool was
ground and quantified with a hemocytometer to estimate
total sporozoite dose. Each bird was inoculated with
1.0 cc of slurry containing approximately 169,000
sporozoites. Six control birds of the same sex and age
were inoculated intraperitonea 1 1 y with 1.0 cc of the
carrier.
On days 3, 5, 8, 11, 14 and 17 post-infection, 1
inoculated poult and 1 control bird were killed by
decapitation. Representative pieces of pectoral muscle,
liver, spleen, lung, heart, brain, kidney, bone marrow,
duodenum, pancreas and cecum were fixed in 10% buffered
formalin and Carnoy's fixative. Tissue fixed in Carnoy's
was dehydrated in absolute ethanol, cleared overnight
in amyl acetate and embedded in paraplast. Sections were
cut at 4 um and stained with hematoxylin and eosin or
Giemsa-colophonium (Bray and Garnham, 1962).


142


282
Oesser, S.S. 1972b. The fine structure of the ookinete of
Pa r a h a ernop roteus ve 1 an s ( =Ha emop roteus v e 1 a n s )
(Ha emospot idia: Ha emop rotei d a e ) Ca n J ^ Zo o 1 .
50: 477-480.
Desser, S.S. 1972c. The fine structure of Leucocytozoon
simondi. V. The oocyst. Can. J. Zool. 51 707-711.
Des ser, S.S., and F. Allison. 1979. Aspects of the
sporogonic development of Leucocytozoon tawaki of the
Fiordland Crested Penquin in its primary vector,
Austrosimulium u n g u 1 a turn: an ul t rast ructura 1
study. Ti Parasitol. 65: 737-744.
Desser, S.S., and A.M. Fa 1 lis. 1967. The cytological
development and encapsulation of mega 1 oschizonts of
Leucocytozoon simondi. Can. J. Zool. 45: 1061-1065.
Desser, S.S., A.M. Falls and P.C.C. Garnham. 1968.
Relapses in ducks chronically infected with
Leucocytozoon simondi and Parahaemoproteus net tionis.
Can. J. Zoo 1. 46: 231-285.
Desser, S.S., and K.A. Wright. 1968. A preliminary study of
the fine structure of the ookinete, oocyst and
sporozoite formation of Leucocytozoon simondi
Mathis and Leger. Can. J. ZooH 303-307.
Dubey, J.P., T.P. Kistner and G. Callis. 1983. Development
of Sarcocyst i s in mule deer transmitted through
dogs and coyotes. Can. J. Zool. 61: 2904-2912.
Duncan, D., J. Eades, S.R. Julian and D. Micks. 1960.
Electron microscopic observations on malarial oocysts
(P1 a smodium ca themerium). J. Protozool. 7: 18-26.
Dyce, A.L. 1969. The recognition of nulliparous and
parous Cu 1 i c oid e s without dissection. J. Aust.
Entorno 1 Soc. SI TT-15.
Entzeroth, R. 1983. Electron microscope study of
merogony preceding cyst formation of Sarcocystis
sp. in Roe Deer. Z. Parasitenkd. 69: 447-456.
Eve, J.H., F.E. Kellog and R.W. Bailey. 1972. Blood
parasites in wild turkeys of eastern West
Virginia. J. Wildl. Manage. 36: 624-627.


154
(Table 15). However, average cell width was greater in
mi ergametocyts from the pheasant and probably related
to the large, lateral displacement oE the host cell nucleus
(Figure 65). The average adjusted number of pigment
granules in mi crogametocytes from the Chuckar was
considerably higher than values from the pheasant and
turkey.
A discriminant analysis was performed on adjusted
variables from a calibration data set derived from 15
m i crogametocytes from each host. It correctly
identified 73.3% of the discriminant scores from the
Chuckar, 60.0% of the scores from the pheasant and 66.7% of
the scores from the turkey (Table 16).
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 microgametocytes
from each of the 3 host species. It correctly identified 3
(75%) Chuckar scores, 3 (75%) turkey scores, but no (0%)
pheasant scores. Two pheasant scores were identified as
turkey, 1 as Chuckar and 1 failed to meet the criteria for
inclusion into any of the 3 categories (Table 17).
Infected host cells macrogametocytes. Host cells
infected with macrogametocytes underwent a number of common
changes in each host species (Figures 60, 62, 64). All had
greater cell lengths and cell areas than corresponding
uninfected cells from the same host species. Infected host


216
1971; Falls and Bennett, 1961; Miltgen et al., 1981). The
relatively low (50%) infectivity of salivary gland
sporozoites ftom specimens of edeni, hinmani and
C. arboricola as well as the lack of infectivity of
sporozoites from the single specimen of haematopotus,
may be related to the problems inherent in handling sticky
salivary glands that are only 100 200 um long.
Another possibility is that the sporozoites observed in
these wild-caught specimens of Cu 1 icoides belonged to
another species of haemosporid i an Bennett and Coombs
(1975) reported a sporozoite prevalence of 13.6% in 184
individuals of sti lobezziodes f r om insular
Newfoundland. However, only 1 individual (0.5%) out of 210
unengorged, wild-caught specimens of ed e ni from
Paynes Prairie was positive for salivary gland sporozoites
by dissection, in spite of the fact that collections
were made during periods of active transmission of H.
me 1 eagridis. Natural infections were not detected in
salivary gland dissections of unengorged specimens of C.
arbor icola or C_^ h i nma n i Attempts to transmit H.
me 1eag ridis with individual Cu 1 icoi des and with pools of
several specimens that had been ground in a suitable
carrier were more successful.
The 17 day prepatent period of successful experimental
infections was less than the 28 day period that Greiner and
Forrester (1980) estimated from observations of sentinel


104
and brownish-black. Those in the lung were smaller and
golden-brown. All were contained in macrophages.
Follicular hyperplasia was corrmon in the spleens of all
infected birds. The degree of hyperplasia as well as
the number of pigment deposits varied directly with the
size of the infective dose.
Sections of gut from the control, low dose and high
dose birds had a few multifocal areas of infiltrate
composed of heterophils. Coccidian parasites were not
observed. One high dose bird had a granulomatous
peritonitis. The granulomas were composed of macrophages,
heterophils and giant cells that surrounded amorphous,
eosinophilic masses.
Sections of brain, bone marrow, kidney, heart,
proventri cu 1 us and gizzard were unremarkable.
Parasitemia. Young gametocytes appeared in the
peripheral circulation of all birds in the high and low
dose groups at 17 DPI. The control birds remained
uninfected throughout the study. The parasitemia in
both infected groups quickly reached a peak by 21 DPI
and then rapidly fell within 7 days to values less than 10%
of those at the crisis (Figure 36). A second, smaller peak
occurred at approximately 5 weeks pos t-infection Both
groups remained patent through the course of the study,
although parasitemias were often less than 1.0 %.


231
e g. Toxoplasma and Hammondia, develop in other organs
or tissues (Frenkel and Dubey, 1975; Frenkel, 1973).
The single, 3-day-old schizont that was found appeared
to be within the capillary endothelium (Figure 42).
However, the presence of disrupted muscle fibers in the
infected bird and the development of 5-day-old schizonts
both within and between muscle fibers suggests that
development may occur in other locations as well
(Figures 43, 44, 53).
Studies of other avian haemoproteids have reported
development of the mature exoerythrocytic stages in
capillary endothelial cells of lung, liver, spleen, heart,
kidney and cecum (Khan and Fallis, 1969; Ahmed and
Mohammed, 1977; Bradbury and Gallucci, 1971; Sibley and
Werner, 1984; 3aker, 1966a; O'Roke, 1930; Greiner, 1971).
Farmer (1965 ) found large mega 1 oschizonts within the
gizzard muscles of Rock Doves with natural infections of H.
sacharovi. He suggested that they may be the
exoerythrocytic stages of H^ sacharovi, but was unable
to transmit the infections with transplants and injections
of infected gizzards. Similar forms have not been found in
Mourning Doves infected with HL sacharovi (Farmer, 1965;
Greiner, 1971) and most workers have ignored the
possibility that they may be part of a haemoproteid life
cycle. Mi 1tgen et al. (1931) reported the development
of megaloschizonts within muscle tissue of


Figure 78. Extracellular exf1 age 1lating mi ergametocyte.
Axonemes (A) in various stages o£ assembly are
associated with arm-like extensions (double
arrows) of the gametocyte nucleus (N) The
gametocyte has a cytostome (large arrow),
mitochondria (M) and a 2-layered pellicle with
a discontinuous inner layer. X 46,400.
Figure 79. Extracellular exf1 age 11 ating mi ergametocyte.
The central microtubules (arrows) of the
transversely sectioned axoneme have periodic
striations along their length. X 66,700.


65
Table 10. Attempted isolations of Haemoproteus me 1 eagridis
from pools of Cu I icoi dFs h i nman FT Cu i i co i de s
a r bo rico1 a and Cu 1 icoi des knowltoni at Paynes
Prairie PAP) and Fisheating Creek (FEC).
Cu 1 icoi des hinmani
Location
Month
Year
Pool s
it Flies
Isolations
PAP
July
1983
1
3
0
PAP
August
1983
1
4
0
PAP
September
1983
1
8
0
FEC
May
1983
1
10
0
FEC
Augus t
1983
1
2
0
FEC
March
1984
1
7
0
FEC
April
1984
I
5
0
Culicoides arboricola
Location
Month
Year
Pool s
it Flies
Isolations
PAP
April
1983
1
10
0
PAP
May
1983
1
13
0
PAP
Ma r c h
1984
1
5
0
FEC
April
1984
1
8
0
Culicoides knowltoni
Location
Month
Year
Pool s
it Flies
Isolations
FEC
June
1983
1
3
0
FEC
July
1983
1
4
0
FEC
September
1983
1
4
0
FEC
May
1984
2
53
0


Figure 39. Average hematocrits for high dose birds (
low dose birds ( A A ) and control birds
There were no statistically significant di
among groups when comparisons were made by
0.0S).
O O ),
( ).
fferences
week (p


37
for 4 weeks following inoculation. They were fixed and
stained as described earlier.
The slurry from a second pool of 37 engorged
specimens of eden i and 84 engorged specimens of C.
hinmani, collected in July, 1984, was divided equally
among 2, 3-day-old Northern Bobwhites, Col inus virginianus,
2, 7-day-old Rhode Island red chickens, Ga11u s ga 1 1 us ,
and 2, 5-day-old broad-breasted white turkeys. The
chickens were obtained from a local hatchery and the quail
were obtained from the Department of Poultry Science,
University of Florida. Each bird was inoculated IP with
0.1 cc of slurry containing approximately 5,000
sporozoites. A pair of uninfected birds of each species
was kept as negative controls. All birds were maintained
and bled as described earlier. Tissues from birds that
died prior to the end of both experiments were fixed in
10% buffered formalin and embedded, sectioned and stained
as described earlier.
Morphometric Analysis
Parasitemias were quantified as the number of
parasites per 10,000 red blood cells as described earlier.
Morphometric parameters were determined from a maximum
of 15 mature, 7- to 9-day-old mi crogametocytes and 15
mature, 7- to 9-day-old macrogametocytes from each host.
Measurements were made by the methods of Bennett and
Campbell ( 1972) as modified by Forrester et al (1977).


Figure 33. Nodular infiltrate from pectoral muscle of a
high dose bird that was killed at 8 weeks
post-infection. The infiltrate is composed
primarily of mononuclear cells and surrounds
some dark-sta in i ng areas of calcification
(arrow). Hematoxylin and eosin. Bar = 50 um.
Figure 34. A thrombus, composed of fibrinous material,
adjacent to the remnants of a degenerating
mega 1oschizont (arrow). The section is from
pectoral muscle of a high dose bird that was
killed at 8 weeks post-infection Hematoxylin
and eosin. Bar = 20 um.
Figure 35. A mass of degenerating muscle fibers,
infiltrated with macrophages and heterophils.
Section is from pectoral muscle of a high dose
bird that was killed at 8 weeks
post-infection. Hematoxylin and eosin. Bar
= 20 um.


177
electron density and were often indistinguishable in areas
where the basement membrane was stretched tightly over the
oocyst wall (Figure 83).
The pellicle surrounding the 3-day-old sporo'olast was
imme diately interior to the oocyst wall. It was
composed of a single unit membrane, underlaid in many
places with dark, osmiophilic thickenings (Figure 83).
Prominent, membrane-bound, lipid-like inclusions were
clustered together near the center of 3-day-old oocysts
(Figure 83). Surrounding these and scattered throughout
the cytoplasm were numerous mitochondria with tubular
cristae (Figure 83). Several irregularly shaped nuclei
with prominent nucleoli were present around the
periphery of the parasites. The nucleoplasm had a density
similar to the cytoplasm and was difficult to discern in
areas where the nuclear envelope was indistinct (Figure
83). Spindle fibers and kinetic centers were not observed.
Six-day-old oocysts. 3y 6 days, oocysts ranged in
development from immature forms, resembling 3-day-old
oocysts, to mature oocysts that had ruptured and
released their sporozoites. Oocysts that were more mature
than the 3-day-old forms had more space between the
sporoblast body and the oocyst wall. Numerous budding
sporozoites developed around the periphery of the
sporoblast. The buds originated under the osmiophilic




161
Table 15. Average adjusted measurements of
mi ergametocytes. Average measurements of each
variable are expressed as a percentage of the
average uninfected host cell area.
Variable Chuckar Pheasant Turkey
Paras ite
Length
41.4
(1.96)+
32.2 (5.65)
37.5
(6.01)
Paras i te
Width
4.6
(0.67)
6.1 (1.21)
5.8
(0.67)
Parasite
Area
103.7
(9.03)
86.1 (8.38)
106.3
(10.6)
P i gme n t *
39.5
(12.0)
25.3 (5.88)
24.3
(4.72)
N =
15
15
15
+ Standard deviation
* Adjusted average of number of pigment granules.


Figure 13. Transmission vs. abundance of 3 species of Cu 1 icoi des at
Site A at Paynes Prairie that were able to support
development of Haemoproteus me 1eagridis (-- =
Bennett trap catch; Q.O ~= New Jersey light trap
catch). The kite diagram at the top of the figure
indicates the % of sentinel birds that became
infected with Haemoproteus me 1eagridis during 4-week
periods between May, 19 8 2 and June, 1 984 Blank
areas in the diagram indicate periods where transmission
did not occur. Marks on the x-axis that follow each
month indicate the middle of that month.


233
Failure to detect developing schizonts at 11 days
post-infection may have been because the infected bird had
an infection of low intensity. The large difference in
size between the 8-day-old and the 14-day-old forms
makes it unlikely that a third generation of schizogony
could have taken place in other tissues.
The large size of the second-generation
megaloschizonts of me 1e a a ridis is comparable to
mega 1oschizonts that have been found in birds with natural
infections of aa r nhami desser i and s acharovi
(Garnham, 1966; Miltgen et al., 1981; Farmer, 1965).
However, the megaloschizonts of aarnhami contained
cytomeres, separated from each other by distinct septa,
lacked a thick, hyaline wall and have not been found in
muscle tissue (Garnham, 1966). Miltgen et al. (1981)
described "pseudo-septa" in inmature megaloschizonts of H.
de s s e r i The authors suggested, though, that these were
artifacts caused by contraction of the surrounding
muscle fibers during fixation of the tissue. The presence
of a hyaline cyst wall around megaloschizonts of H,
de s se ri, the absence of septa in mature forms and the
exclusive development in muscle tissue were very similar to
H. me 1eag ridis. The thick-walled, aseptate cysts
described by Farmer (1965) were also very similar to mature
megaloschizonts of me 1eagridis, but their development
appeared to be limited to gizzard muscle. Immature


192


238
suggesting that the early inflanmatory reaction may have
been to dead and dying host cells and parasites.
The host response surrounding 14-day-old and
17-day-old second-generation mega 1oschizonts is similar to
the host reactions to the large intramuscular cysts of
Sarcocystis. Leek et al. (1977) noted multifocal
perivascular inflamnatory infiltrates in the musculature of
lambs that were experimentally infected with Sarcocys tis.
The response was not associated with developing sarcocysts,
but was apparent around degenerating cysts. Mundy et al.
(1975) reported similar results. Other studies of
Sarcocystis and Hammondia have reported the necrosis and
mineralization of muscle fibers adjacent to developing
cysts (Cawthorn et al., 1984; Frenkel and Dubey, 1975).
Among haemosporidian parasites, the host response
to mega 1 oschizonts of me Ieagridis is similar to host
reactions to simondi. Miller et al. (1983), Desser
(1967), Cowan (1957) and Newberne (1957) reported the
presence of mixed inflammatory infiltrates composed of
mononuclear cells, heterophils, plasma cells and red blood
cells around both intact and ruptured mega 1 oschizonts as
well as necrotic changes in the surrounding host
tissue. Cowan (1957) and Miller et al. (1983) described
the spontaneous necrosis of Leucocytozoon
megaloschizonts in the absence of invading host cells. In
both studies, necrotic mega 1oschizonts were filled with an


36
Host Speci£ icity
Infection of Hosts
In June and July, 1984, domestic turkeys infected
with H^ me 1 eagr i d i s were used as bait birds in Bennett
traps operated at Fisheating Creek. Engorged specimens
of eden i and h i nman i were collected, held at 25o
C for 7 days to allow development of sporozoites, pooled
and ground in Aedes aegyp t i Ringers (Hayes, 1953). The
slurry from the first pool of 40 specimens of edeni
and 32 specimens of hinmani, collected in June, was
divided equally among 2, 2-day-old Chukar Partridges,
A1e c t o ris c h u c k a r, 2, 2-day-old Guineafowl, Numid a
me 1eagr i s 2, 7-day-old Ring-Necked Pheasants, Phasianus
co1chicus and 2, 7-day-old broad-breasted white turkey
poults. The Chuckars, Guineafowl and Ring-necked Pheasants
were obtained from Morris Hatchery (Miami, Florida).
Each bird was inoculated IP with 0.15 cc of slurry.
Sporozoite counts of the slurry with a hemocytometer
revealed that each bird received approximately 375
sporozoites. Pairs of uninfected Chuckars, Guineafowl,
Ring-necked Pheasants and turkeys were kept as negative
controls. All birds were housed by species in separate
battery cages in a vector-proof room and fed and watered
as described earlier. Smears were prepared from blood
obtained from leg veins of all birds, 3 times a week,


Figure 24.
Figure 25.
A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection .
Hyaline, necrotic muscle fibers (large
arrow) and a mononuclear infiltrate (double
arrow) surround the schizont. Hematoxylin and
eos in. Bar = 50 um.
A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that
spontaneously at 19 days post-infection .
The cyst contains numerous mononuclear cells
(arrow) and a fibrinous exudate. Giant
cells (double arrow) are adjacent to 1 side of
the cyst. Hematoxylin and eosin. Bar = 50 um.


Figure 7. Scatter plot of capture times for specimens of
C u 1 i c o i d e s hinma ni that were captured in Bennett
traps at P a y n e s P rairie and Fisheating Creek.
Capture time is plotted as minutes before or after
nautical sunset (reference line). Arrow indicates mean
capture time.




262
electron dense thickening and associated intranuclear
spindle in 1 macrogametocyte Since serial sections
were not cut, it is likely that they may 'nave been missed.
A single cytostoine normally occurs in mature
gametocytes of columbae and rL metchnikovi (Bradbury and
Roberts, 1970; Sterling, 1972). Gallucci (1974a) and
Bradbury and Trager (1968a) reported "internal cytostomes"
and a single, non-functional peripheral cytostome within
mature and immature macrogametes of fL columbae. The
internal cytostomes consisted of 2 electron dense rings,
but were located in the interior of the cell rather than in
the external pellicle. Other observations of cytostomes in
maturing macrogametes and exflagellating
mi ergametocytes have not been reported. The presence
of cytostomes in developing gametes of me 1eagridis
indicates that they persist as part of the pellicle
throughout development of the gametes. However, their
small size and the absence of associated food vacuoles
indicates that they are non-funct iona 1 .
Microgametogenes i s A number of studies of the
mi crogametogenes i s of avian haemoproteids have been
conducted. Bradbury and Trager (1968a, 1968b), Sterling
(1972) and Aikawa and Sterling (1974a) studied co1umbae
and metchnikovi. They reported the development of
axon ernes composed of 9 peripheral microtubules that


190


175
mi crogametocytes developed breaks and discontinuities
throughout its surface so that the middle layer of the
pellicle became the new, outer limiting membrane of the
parasite (Figures 71, 72). By contrast, the inner layer of
macrogametocytes remained intact (Figure 74).
Macrogametocyte nuclei were elongate and extended
across the diameter of the parasites. A prominent
nucleolus was present at 1 pole. A kinetic center composed
of a mass of amorphous, e 1 ectron-dense material was
adjacent to the nuclear envelope at the other pole (Figure
74).
Within 3 minutes after gametogenesis began,
macrogametocytes and mi ergametocytes were free of their
host cells. Both were bound by the middle layer of the
original 3-layered pellicle. The pale, electron lucent
remnants of ruptured host cell nuclei as well as
remnants of host cell membranes were often adjacent to
or surrounded free gametocytes (Figure 77). Both maturing
macrogametes and exf1 age 1 I ating mi ergametocytes contained
a single cytostome (Figures 76, 78).
Maturing macrogametes were packed densely with
ribosomes and contained numerous mitochondria and an
extensive network of smooth endoplasmic reticulum
(Figure 75). They contained an elongate nucleus with a
prominent nucleolus that extended across the center of the
macrogametocyte (Figure 75).


57
Table 2. New Jersey light trap collections Paynes Prairie
May 1982 July 1984
Site A*
Site Bit
Site A +
B
Species
Total (%)
Total (%)
Total
(%)
C.
i ns ignis
1,501
(14.3%)
17,495
(73.3%)
18,996
(55.3%)
cr
eden i
2,693
(25.7%)
825
(3.5%)
3,518
( 10.2%)
cr
s t e 1 1 i f e r
1,474
(14.1%)
1,513
(6.3%)
2,987
(8.7%)
cr
arborico1 a
1,706
(16.3%)
1,070
(4.5%)
2,776
(8.1%)
cr
crepuscu laris
915
(8.7%)
475
(2.0%)
1,390
(4.1%)
CT
spinosus
184
(1.8%)
919
(3.9%)
1,103
(3.2%)
CT
scan 1 oni
225
(2.1%)
471
(2.0%)
696
(2.0%)
cr
nanus
606
(5.8%)
73
(0.3%)
679
(2.0%)
CT
deb i 1 i pa 1 pus
416
(4.0%)
90
(0.4%)
506
(1.5%)
CT
n i ger
120
(1.1%)
363
(1.5%)
483
(1.4%)
CT
v i 11 os ipennis
152
(1.5%)
106
(0.4%)
258
(0.8%)
CT
h i nina n i
148
(1.4%)
95
(0.4%)
243
(0.7%)
CT
paraensis
130
(1.2%)
41
(0.2%)
171
(0.5%)
cr
ousairani
61
(0.6%)
39
(0.2%)
100
(0.3%)
CT
venustus
5
(<0. 1%)
89
(0.4%)
94
(0.3%)
CT
bauer i
33
(0.3%)
37
(0.2%)
70
(0.2%)
CT
bick1eyi
25
(0.2%)
45
(0.2%)
70
(0.2%)
CT
haematopotus
40
(0.4%)
23
(0.1%)
63
(0.2%)
CT
a 1achua
I
(< 0. 1%)
44
(0.2%)
45
(0.1%)
CT
biguttatus
19
(0.2%)
19
(0.1%)
38
(0.1%)
CT
t i s sot i
12
(0.1%)
18
(0.1%)
30
(0.1%)
CT
gut tipennis
8
(0.1%)
9
0. 1%)
17
(0.1%)
C. pechumani
3
(< 0.1%)
0
(0.0%)
3
(< 0. 1%)
CT
furens
0
(0.0%)
2
(<0. 1%)
2
(<0. 1%)
C. mulrennani
0
(0.0%)
2
(<0. 1%)
2
(<0. 1%)
CT
chiopterus
1
(< 0.1%)
0
(0.0%)
1
(<0. 1%)
CT
pi 1ifetus
0
(0.0%)
1
(<0. 1%)
1
(<0. 1%)
Total 23,864 10,478 34,342
* 52 nights of trapping
it 44 nights of trapping


Figure 84. Six-day-old oocyst from a specimen of
Cu 1 ic o id e s e d e n i Budding sporozoites
(large arrows) are spaced around the periphery
of the sporoblast body. Large nuclei (N)
are located are located at the periphery of the
sporoblast body, beneath the budding
sporozoites. The remnants of the electron
dense apical complex of the ookinete (Ac),
as well as large, lipid-like inclusions (L),
are near the center of the oocyst. X 26,100.
Figure 85. Six-day-old oocyst from a specimen of
Cu 1icoi des edeni. Budding sporozoites contain
e1ectron dense rhoptries (R), a nucleus (N), a
polar ring (small arrow) and subpe1 1 i cu 1 ar
microtubules (large arrows). The outer
layer of the pellicle that surrounds the
sporozoites is underlaid by 2 unit membranes in
close apposition to one another (double
ar row). X 78,300.


THE EPIZOOTIOLOGY AND PATHOGENICITY OF
Haemopco me ^eagH cH s Levine, 1961,
BY
CARTER TAIT ATKINSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1985


215
meleagridis Cu 1 i c i od e s crepuscularis is related
closely to knowl ton i and replaces it in northern Florida
and throughout Nortii America (Blanton and Wirth, 1979). It
is the only species of Cu I i co i des reported from Florida
that has been implicated previously as a vector of avian
haemoproteids (Bennett and Falls, 1960). However,
specimens of crepuscularis were collected rarely in
Bennett Traps and light traps at Paynes Prairie and it
is unlikely that this species plays an important role in
the natural transmission of H_^ meleagridis in Florida.
Specimens of h a ema t opo t u s were captured too
infrequently at Paynes Prairie and Fisheating Creek to
implicate this species as a vector of meleagridis at
either site. However, it is related closely to C^ edeni
(31anton and Wirth, 1979). Since the distribution of C.
eden i is limited to Florida and the Bahamas (Blanton and
Wirth, 1979), C^ haematopo tus may be an important vector in
other parts of North America where the prevalence of H.
me 1eagridis is high.
Sporogonic Stages and Experimental Transmission
The size and morphology of the ookinetes, oocysts and
sporozoites of me 1 eag ridis, as well as the relatively
short sporogonic cycle of 6 7 days, resemble the findings
for other species of Haemop rot eu s known to develop in
ceratopogonids (Falls and Wood, 1957; Khan and Falls,


255
Haemoproteus stab1er i can be distinguished from
me 1e a 3 cidis and man son i by the amoeboid margins of
mature gametocytes, their significantly fewer pigment
granules and the higher host cell nuclear displacement
ratio (White and Bennett, 1979; Greiner and Forrester,
1980).
The strong morphological similarities between
h a emop r o t e i d s of Grouse and me I e a g r i d i s may be
significant. Bennett (1960) suggested that the host
behavior, habitat and corrmunity may be more important in
determining the host range of avian haemoprote i ds than
specificity of the parasites themselves. Since the
distribution of Ruffed Grouse coincides with that of the
Wild Turkey throughout the Appalachian Mountains of the
eastern U.S., transmission of fR mansoni and rR me 1eagridis
between either host may be possible. Fall is and Bennett
(I960) reported the sporogony of man soni in species
of Cu 1icoi des from Canada. It is likely that species of
C u 1 i c o i d e s are vectors of man son i throughout the
range of its host. Studies of the experimental cross
transmission of rK me 1 e agridis and man son i between
Ruffed Grouse and turkeys are needed to determine the
validity of both species of Haemoproteus
The results of the morphometric analysis of H.
meleagridis in turkeys, the Ring-necked Pheasant and the
Chuckar demonstrate that few significant changes in


RESULTS
Epizooti o 1ogy
Vectors
Paynes Prairie. During 96 nights of trapping between
May, 1982, and July, 1984, 34,342 specimens of
Cu 1icoides belonging to 27 species were captured in New
Jersey light traps at Sites A and B (Table 2). Cu 1 i co i des
i n s i g n i s was the most common species taken at Paynes
Prairie. However, 79% of the total catch of 18,996
individuals was captured in a single night at Site B in
November, 1982. Specimens of 9 additional species, i.e. C.
e d e ni ( 10.2%), stellifer (8.7%), arbor ico1 a
(8.1%), crepuscular i s (4.1%), spinosus (3.2%), C.
s can 1 on i (2.0%), nanus (2.0%), deb i 1 i pa 1 pus (1.5%)
and niger (1.4%), made up approximately 40% of the
remaining catch. Specimens of the other 17 species were
captured infrequently or in low numbers.
Of the specimens of 27 species of Cu 1 icoi des captured
in New Jersey light traps, only representatives of 12
species, totaling 1,428 engorged individuals, were taken in
42


Figure 86. Six-day-old oocyst from a specimen of
Cu 1icoi des edeni. Budding sporozoites (arrows)
contain electron dense rhoptries (R). As they
complete their maturation, a small residual
body containing large, lipid-like vesicles (L)
rema ins. X 34,800.


38
Fifteen uninfected red cells were also measured from each
host. All measurements were made from camera lucida
drawings of infected and uninfected erythrocytes that
were adjusted to the proper scale with a slide micrometer.
A discriminant analysis was performed on measurements
of mi crogametocy t e s mac r ogame t ocy t e s and infected host
cells f r om each species susceptible to fU me 1 eagr i d i s
to determine whether parasite and host cell morphology
differed in each host. Since uninfected red cells from
each host species differed in size and could have a
limiting effect on parasite size, measurements of parasites
were expressed as a percentage of the average area of
uninfected red cells from the same host. To standardize
morphological changes in host cells infected with
gametocytes, measurements from infected host cells were
expressed as a fraction of the corresponding average
measurement of uninfected host cells of the same species
(e.g. infected host cell area/average host cell area).
Before analysis, all variables were tested for normality
with the Shapiro-Wilk statistic (SAS User's Guide: Basics,
1982). Since approximately 1/4 of the variables were
non-normal, a nonparametric, nearest neighbor analysis
(k = 3) was performed (SAS User's Guide: Statistics, 1982).
Adjusted measurements of macrogametocyte length and
width, nucleus length and width, gametocyte area, nucleus
area and number of pigment granules were used as predictor
variables to generate a discriminant function based on


284
Freund, R.J., and R.C. Lit tell. 1981. SAS for linear
models. SAS Institute Inc., Cary, North Carolina.
231 pp.
Gabaldon, A., and G. Ulloa. 1980. Ho 1oendemicity of
malaria: an avian model. Trans. Roy. Soc. Trop.
Med. Hyg. 74: 501-507.
Gallucci, B.8. 1974a. Fine structure of Haemoproteus
columbae Kruse during macrogametogenesis and
fertilization. J. Protozool. 21: 254-263.
Gallucci, B.B. 1974b. Fine structure of Haemoproteus
co1umbae during differentiation of the ookinet e. T7
Protozool. 21: 264-275.
Gardiner, C.H., H.J. Jenkins and K.S. Mahoney. 1984.
Myositis and death in Bobwhites, Col in u s
virginianus (L.), due to hemorrhagic cysts ol a
h a emo sporozoan of undetermined taxonomic status.
J. Wi Id. Dis. 20: 308-318.
Garnham, P.C.C. 1951. The mosquito transmission of
Plasmodiurn inui Ha 1berstaedter and Prowazek, and its
pre-erythrocytic development in the liver of the
Rhesus monkey. Trans. Roy. Soc. Trop. Med. Hyg. 45:
45-52.
Garnham, P.C.C. 1966. Malaria parasites and other
Haemosporidia. Blackwell Scientific Publ., Oxford.
1114 pp.
Garnham, P.C.C. 1973a. Epizootics of Leucocy tozoon
infections in parakeets in England, in "Progress fn
Protozoology". Univerit de Cler mo n t C1 e r mo n t ,
France, p. 149.
Garnham, P.C.C. 1973b. Unusual hosts for parasites
under natural and experimental conditions, in
"Proc. Fourth Internat. Cong. Protozool.". Universit
de Clermont, Clermont, France, pp. 193-194.
Garnham, P.C.C., R.G. Bird, J.R. Baker, S.S. Desser and
H.M.S. El-Nahal. 1969. Electron microscope studies on
motile stages of malaria parasites VI. The ookinete of
Plasmodium berghei yoe 1 i i and its transformation into
the early oocyst. Trans. Roy. Soc. Trop. Med. Hyg.
63: 187-194.


Figure 19. Departures from normal for average monthly
temperatures during 1982, 1983 and 1 984- at
Fisheating Creek. Normals are based on 30 year
averages each mo nth between 1951 and 1980
(Climatological Data: Florida, 1982, 1983,
1984).


228
was still high, suggests that the inhibition of
sporogony by lower environmental temperatures may be 1
factor limiting the winter transmission of H.
me 1eag r i d i s at Paynes Prairie. During this study,
biting activity was never detected for any species of
Cu 1 ico i d e s below ISo C. Inhibition of host-seeking
activity by cool evening temperatures between January
and March may also be a major factor in restricting the
winter transmission of me 1eagridis.
The ep i zooti o 1ogy of fh meleagridis in southern
Florida has many similarities to stable, holoendemic
malaria (Wernsdorfer, 1980; Gabaldon and Ulloa, 1980). At
any given time during the year, most of the Wild Turkey
population at Fisheating Creek is infected with H.
me 1eagrid i s and has circulating gametocytes, available for
potential vectors. The high prevalence of H
me 1eagridis and the large, relatively stable vector ( =
C. edeni) population insure that a high rate of natural
transmission can occur throughout the year. In these
respects, the epizootiology of this parasite is
significantly different from other avian haemosporid i ans
that have been studied in temperate North America. In
temperate locations, populations of vectors as well as the
prevalence of avian haemosporidians are highly
seasonal. It has been documented that the spring
relapse of Haemop r o t eus PI asmod i um and Leucocy tozoon i s


254
Four species, fh chukari, H. coturnix, sal 1 inarum and H.
perdix, were reported as nomen nuda. Bennett et al.
( 1 982 ) lists ba 1 four i from Guineafowl as valid, but
descriptions and figures of tilis species are absent from
the cited references. It will also be considered a
nomen nudum. Under the current system of haemoproteid
taxonomy, these species must be morphologica11y distinct
from one another to retain their status as separate taxa.
Haemoproteus me 1eagridis i most similar to rU man soni
and 1-L s t a b 1 e r i of Grouse (subfamily Tetraoninae).
Gametocytes of other species of Haemoproteus from the
subfamilies Phasianinae and Numidinae are halteridial
and only partially encircle the host cell nucleus. The
geographical range of ma n s o ni over laps mo s t of the
geographical range of me 1eagridis, while rh stableri has
been reported only from Ruffed Grouse in Montana (White and
Bennett, 1979). Greiner and Forrester (1980) discussed the
similarities between rh mansoni and rh meleagridis and
stated that the number of pigment granules in circumnuc1 ear
gametocytes of mansoni was significantly fewer than
in circumnuc1 ear gametocytes of meleagridis. They also
noted that the number of pigment granules in circumnuclear
forms of mansoni was fewer than in halteridial
forms. By contrast, the number of granules increased as
gametocytes of fh me 1eagridis became circumnuc1 ear.


267
Allison, 1979). Sterling and DeGiusti (1974) observed
these organelles throughout the development of oocysts
of metchnikovi and in the residual body after sporozoite
differentiation was complete. Apical organelles derived
from the ookinete of meleagridis appear to persist In
the cytoplasm throughout the life of the oocyst. There is
no evidence that any of them are reused in the formation of
sporozoites (Garnham et a 1., 1969).
The early stages of sporozoite formation in PIasmodium
are initiated by vacuolization of the peripheral oocyst
cytoplasm. This process also occurs in early oocysts of H.
metchnikov i but has not been described in any of the 3
species of Leucocytozoon that have been studied. Terzakis
et al. (1966) suggested that the vacuoles inay develop after
a local change in ionic concentration in the peripheral
cytoplasm causes water to cross the oocyst wall or,
alternatively, after the oocyst cytoplasm secretes fluid at
the oocyst periphery. In support of the latter, Sinden and
Strong (1973) described the formation of vacuoles in
oocysts of falciparum by the fusion of membranous
vesicles that originated in the oocyst cytoplasm.
Electron dense, linear thickenings appear under the
membrane soon after vacuolization begins. The vacuoles
eventually coalesce to form cytoplasmic clefts which
subdivide the oocyst cytoplasm. Sinden and Strong
(1978) found that clefts in the oocysts of falciparum


LIST OF FIGURES
Figure Page
1 Ookinete of Haemoproteus me 1eagridis 67
2 Developing oocysts of Haemoproteus me Ieagridis
on the midgut of a specimen of CUTicoides
eden i 67
3 A 6-day-old, degenerating oocyst of
Haemoproteus me 1eagridis from a specimen of
Cu 1icoi des edeni 67
4 A mature, 6-day-old oocyst of Haemoproteus
me 1e a g ridis from a specimen of Cu I icoides~edeni 67
5 One of the 2 salivary glands from a specimen
of Cu 1 icoi des edeni 67
6 A crushed salivary gland from a specimen of
Cu 1 icoi des edeni 67
7 Scatter plot of capture times for specimens of
Cu 1 i co i des li i nrnan i 69
8 Scatter plot of capture times for specimens of
Cu 1 icoi des edeni 71
9 Scatter plot of capture times for specimens of
Cu 1 icoi des arbor ico1 a 73
10 Scatter plot of capture times for specimens of
Cu 1 icoi des knowltoni 75
11 Modified New Jersey suction trap catches of
specimens of Cu I icoi des edeni at Fisheating
Creek 77
12 Modified New Jersey suction trap catches of
specimens of Cu 1 icoi des hinma ni at Fisheating
Creek 79
x




34
von Kossa's stain (Humason, 1979). Pectoral muscle from
2 of these birds was fixed and processed for electron
microscopy as described later.
Data on weight, tarsometatarsal length, hematocrit,
plasma protein concentration and hemoglobin were analyzed
with the SAS general linear models procedure as a
split-plot design with subjects as main plot units and
subjects at a particular time as a subplot unit (Freund
and Littel, 1983). Treatment, subject(treatment), week
and treatmentweek were tested for each variable using
the Type 111 sum of squares. A p value of 0.05 or smaller
was considered significant. When treatment*weeks
interactions were significant, further comparisons were
made by treatment and by week with Duncan's Multiple Range
Test. A comparison of organ weights at necropsy was done
with a one-way analysis of variance using the SAS general
linear models procedure (SAS User's Guide: Statistics,
1982).
Experiment 2 Exoerythrocytic Development
A second series of experimental infections was
conducted to study the development of the early
exoerythrocytic stages and their associated host response.
Four domestic turkeys, infected as sentinel birds at
Fisheating Creek, were exposed in Bennett traps at
Fisheating Creek in May, 1985, on days 8-10 of patency.


62
Table 7. Mean capture times for specimens of Cu 1 ico i des
taken in Bennett traps at Paynes Prairie and
Fisheating Creek. Values are in minutes,
relative to nautical sunset. Numbers in
parentheses are standard deviations.
Cu 1 i coi des Species Paynes Prairie
Quarter*
hinma ni
eden i
a rborico1 a
1
18.7a
26.3a
(26.9)
(22.0)
2
-35.3a
15. lb
8.1c
(28.1)
(21.4)
(31.2)
3
-28.0a
12.0b
17.3b
(30.8)
(27.9)
(20.0)
4
-24.6a
14.8b
40.4c
(24.1)
(22.5)
(20.4)
Cu 1 icoi des
Species -
Fisheating Creek
Quarter
hinma ni
eden i
a rboricol a
knowltoni
1
-33.3a
5.1b
21.7bc*
51.5c*
(22.8)
(29.1)
(9.1 )
(4.9)
2
-58.6a
0.0b
21.3c
26.9c
(32.9)
(35.3)
(31.1)
(18.2)
3
-49.3a
6.4b
24.8b*
26.5b
(27.6)
(32.9)
(41.4)
(16.8)
4
-23.9a
14.5b
S5.0c*
31.7bc
(27.8)
(29.4)
(19.1)
(20.4)
a Values with the same letter are not significantly
different, P< 0.05
* N 10
+ Quarter 1 = January-March; Quarter 2 = April-June;
Quarter 3 = Ju1y-Septembet; Quarter 4 = October-December


8
found in certain species of birds as well as the
specificity of the parasites themselves.
Anomalies in the prevalence of columbiform
haemoproteids have led a number of other authors to
question the importance of hippoboscid flies as vectors.
Huff (1932), Hanson et al. (1957) and Greiner (1975) felt
that the low prevalence of hippoboscids on columbiform
birds at times when the incidence of Haemoproteus is high
suggested the involvement of another vector. Greiner
(1975) collected haematopotus and crepuscularis
feeding on Mourning Doves, Zenaida macroura, in Nebraska
during periods of active transmission of Haemop roteus
He hypothesized that they may play important roles as
vectors of macea 1 1 umi and H^ sacharovi. By contrast,
other workers have found a close correlation between the
seasonal occurrence of co1umbae and its hippoboscid
vector, Pseudo 1ynchi a caar iensis, in populations of Rock
Doves, Co 1umba 1 ivia in Michigan (Klei and DeGiusti,
1975). Ayala et al. (1977) found large numbers of
Microlynchia p u si 1 1 a and S ti 1 borne topa podos t y 1 a on
Columbian Eared Doves, Zenaida auriculata, and felt that
these hippoboscids could account alone for the high levels
of transmission of fU macea 11umi they observed throughout
the year. Unfortunately, critical experiments to test
whether columbiform haemoproteids can be transmitted by
species of Cu 1 icoi des have not been done.


108
When comparisons were made within the control group, a
considerable amount of overlap was detected.
Significant differences were not detected among 2, 3, 4, 6
and 7 weeks PI, among 1, 2, 4, 6, 7 and 8 weeks PI and
among 1, 2, 5, 6 and 8 weeks PI. Average values at week 0
were significantly lower than those at any other week PI.
Hemog1obin Statistical analysis of hemoglobin
data revealed that all 4 variables in the model
statement were highly significant (p = .0001).
When comparisons were made by week, high dose birds
had significantly lower average hemoglobin values at 4
weeks PI than either low dose or control birds. No
differences were significant at other weeks PI.
When comparisons were made within the low dose group,
average values at 6, 7 and 8 weeks PI were significantly
lower than the average at week 0. Average values at week 0
were significantly lower than those at 3 and 4 weeks
PI. Average values at 3 and 4 weeks PI were significantly
lower than those at 1, 2 and 5 weeks PI (Figure 41).
Comparisons within the high dose group had
considerable overlap. Average values at 4, 6, 7 and 8
weeks PI were significantly lower than those at 1, 3 and 5
weeks PI. Average values at 0, 3 and 7 weeks PI were
significantly lower than those at 1, 2 and 5 weeks PI.
Significant differences were not detected among 0, 3 and 7


INTRODUCTION
The protozoan genus Haemopcoteus is composed of over
130 different species of blood and tissue parasites of
birds and some reptiles (Levine and Campbell, 1971; Bennett
et al., 1982). Since the discovery of these organisms
by Danilewsky (1889) nearly a century ago, little beyond
the basic a 1 pha-taxonomy of this group has been studied.
This is primarily because laboratory studies of these
species are difficult to accomplish. Transmission of
avian haemoproteids by blood inoculation is rarely
successful because the parasites do not undergo asexual
schizogony in circulating erythrocytes. Attempts by a
number of workers to transmit the parasites using tissue
homogenates or transplants containing the exoerythrocyt ic
stages have occasionally succeeded, but have not been
reliable enough for practical use (Gondor, 1915; O'Roke,
1930; Coatney, 1933; Lastra and Coatney, 1950; Bierer
et al., 1959).
Another major obstacle to the study of this genus
has been the absence of a convenient laboratory model.
Most studies of the life cycle, fine structure and
1


221
been made o£ any of the avian species of P1 a smod i um,
Leucocytozoon o r Haeinopcoteus. Those that have been
conducted have relied primarily on repeated surveys of the
host population by blood smears to monitor the
prevalence of the parasites under study (Herman, 1938;
Herman et al., 1954; Janovy, 1966; Bennett and Falls,
1960; K1ei and DeGiusti, 1975; Greiner, 1975). While this
procedure has the advantage of monitoring changes directly
in the host population, it is limited by the
difficulties and biases inherent in capturing wild birds
and diagnosing latent and low intensities of infection. In
addition, the precise time when transmission of a blood
parasite begins and its relationship to vector populations
may be difficult to establish.
Chernin (1952) was one of the first to overcome these
problems by using domestic sentinel ducks to monitor
natural transmission of Leucocytozoon simondi Later
workers, including Fa 11 is and Wood (1957), Fa 11 is and
3ennett (1966) and Herman and Bennett (1976), refined these
techniques in more extensive studies of L^ s imond i and
applied them to the study of Haemoproteus ne11ionis. In
conjuction with surveys of wild hosts, the use of sentinel
birds provides a powerful tool for studying the dynamics of
natural transmission. By decreasing the length of time
sentinels are exposed, it is possible to measure the onset
of transmission to within days. Sentinel studies are


126


9
Pathogen icity
Since the first detailed studies of Haemoproteus
in the early part of this century, many authors have
reported instances where this organism appeared to
debilitate the host. Acton and Knowles (1914), Ad i e
(1924), Coatney (1933) and Markus and Oosthuizen (1972)
each observed a pigeon, heavily infected with columbae,
that seemed weak, anemic and with a poor appetite.
Wasielewski and WUlker (1918) reported 6 fatal infections
of Haemoproteus in the thousand they examined and Becker
et al. (1956) attributed the enlarged, purplish livers
they found in domestic pigeon squabs to infections by
H. sacharovi Many workers have speculated that H.
co1umbae and other avian haemoproteids must be pathogenic
to some extent because peak parasitemias may involve half
of the circulating erythrocytes (Levine, 1961; Garnham,
1966). Yet, there have been no experimental attempts
to measure pathological changes in infected birds, even
among columbiform hosts that can be infected fairly easily
in the laboratory.
Only 1 study has attempted to document the
pathogenicity of Haemoproteus in detail. O'Roke (1930)
found that California Quail, Lopho r tyx cal¡fornica,
infected with fL lophor tyx had variable amounts of pigment
deposited in their lungs, testes, spleens and livers.


186


204


188


by macrophages and giant cells. Regenerating muscle fibers
were corrmon (Figure 55).
Small schizonts with blue-gray cytoplasm and numerous,
irregularly shaped black nuclei were present in
capillaries. The schizonts were sausage shaped, with an
irregular outline and ranged from 5 to 8 um in diameter up
to 28 um in length (Figures 45, 46). Most schizonts
extended only as far as 2 to 3, 4 um serial sections. Some
schizonts were found adjacent to or within the muscle
lesions, although most showed no evidence of an associated
host response.
Hepatocellular atrophy and necrosis were evident in
sections of liver from the infected bird. Numerous large,
10 20 um, intracellular vacuoles were scattered
throughout the tissue. The spleen was enlarged, but
otherwise unremarkable. Other tissues from the infected
and control bird were unremarkable.
El even days. At 11 days post-infection, infected
birds continued to improve. Regenerating muscle fibers,
often surrounded by a monocytic infiltrate, were common in
sections of skeletal muscle from I infected bird. Necrotic
areas containing giant cells and macrophages were rare.
Perivascular nodular infiltrates composed of monocytes and
some heterophils were present. Other tissues from the
infected and control bird were unremarkable.


27
effects were tested with a Duncan's Multiple Range Test.
When fewer than 3 species were compared, a T-test was
used (SAS User's Guide: Basics, 1982).
A pair of New Jersey light traps, modified as
described earlier and baited with a paper envelope
containing 2-3 kg of dry ice, was operated continuously
for 24-36 hours at Fisheating Creek during the March,
April and May, 1983, collecting trips. One trap was
suspended 1 meter above the ground and the second, 7 meters
above it, in the middle level of the canopy. Both traps
were operated without light bulbs to minimize diurnal
and nocturnal differences in their attractiveness to biting
arthropods. Sample bottles from each trap were changed
every 2 hours between 1700 and 0900 hours and every 4
hours between 0900 and 1700 hours. Dry ice was replenished
every 4 hours. Data were plotted by as the average number
of individuals of each species of Cu 1 icoi des captured
per hour of trap time for each sampling period.
Experimental Infections
To test the ability of individuals of various species
of Cu 1 icoide s to support development of me 1eagrid i s ,
infected sentinel turkeys were exposed in the Bennett
trap. Engorged flies were dissected at daily intervals
up to 9 days after a blood meal was taken in 0.85 % (w/v)


Figure
60.
Mac rsametocyte of Haemoproteus meleagridis
from an experimentally infected turkey.
Giemsa. Bar = 10 um.
Figure
61 .
Mi crogametocyte of Haemoproteus meleagridis
from an experimentally infected turkey.
Giemsa. Bar = 10 um.
Figure
62.
Macrogametocyte of Haemoproteus meleagridis
from an experimentally infected C'nuckar.
Giemsa. Bar = 10 um.
Figure
6 3 .
Microsametocvte of Haemoproteus meleagridis
from an experimentally infected Chuckar.
Giemsa. Bar = 10 um.
Figure
64.
Macrogametoevte of Haemoproteus meleagridis
from an experimentally infected Ring-necked
Pheasant. Giemsa. Bar = 10 um.
Figure 65. Microgametocyte of Haemop roteus me 1e ag ridis
from an exper i menta1 T"y infected Ring-necked
Pheasant. Gienisa. Bar = 10 um.


45
on the outer wall of the midguts (Figures 2, 3, 4).
Representatives of 5 species of Cu 1 icoi des, i.e. edeni,
C. arboricola, C. h a ema t opo t u s, C. hinma ni and C.
knowltoni, were able to support complete development of H.
meleagridis and had mature oocysts, packed with
sporozoites, and salivary glands with numerous slender
sporozoites by 6 to 7 days after they had taken blood meals
(Table 6) (Figures 4, 6). Cu 1 icoi des edeni was the
most susceptible species. Almost 2/3 of the engorged
specimens of e d e n i developed salivary gland
infections (Table 6). Four oocysts from edeni were
subspherical and measured 14-16.5 um in length (Mean =
15.4, SD = 1.11) and 12-16.5 um in width (Mean = 14.1, SD =
1.84). Oocysts contained from 50-100 elongate sporozoites
that were aligned parallel to one another. A small,
eccentric residua! body composed, in part, of golden-brown
pigment granules was present in each oocyst (Figure 4).
Salivary gland sporozoites were within secretory cells
of the single major lobe that composed each of the 2 glands
(Figure 5). In fresh preparations, they often flexed
and twisted within the salivary gland. Fifteen
Gi ems a-stained sporozoites from 1 specimen of eden i
measured 9.25-12.5 um in length (Mean = 11.1, SD = 0.8) and
0.5-1.0 um in width (Mean = 0.69, SD = 0.17). The nucleus
was located approximately 1/3 of the total length from one


157
pheasant scores were correctly classified as pheasant
(Table 23).
Fine Structure
Mature Gainetocytes
Mi ergametocytes and macrogametocytes were bound by a
pellicle composed of 3 unit membranes. The innermost
membrane was thicker and more osmiophilic than the outer 2
(Figures 66, 67). Other common organelles included a
nucleus bounded by 2 unit membranes, mitochondria with
tubular cristae and food vacuoles that contained
osmiophilic masses of pigment (Figures 66, 67). Both
macrogametocytes and microgametocytes had granular
nucleoplasm with an electron density similar to the
cytoplasm (Figures 66, 67). Macrogametocytes had an
electron-dense nucleolus (Figure 67).
Mature mactogametocytes were packed with numerous
ribosomes that gave the cytoplasm a granular appearance
(Figure 67). By contrast, mature microgametocytes
contained fewer ribosomes and had a paler, more
amorphous cytoplasm (Figure 66). Rough endoplasmic
reticulum and a Golgi apparatus were not observed.
However, gametocytes had a network of smooth endoplasmic
reticulum that extended throughout the cytoplasm. In


259
moderately electron dense material within the dilated
cisternae o£ the endoplasmic reticulum (Sterling and
Aikawa, 1973; Bradbury and Roberts, 1970). Desser et
al. (1970b) suggested that it was the precursor to the
crystalloid material observed in ookinetes, oocysts and
sporozoites of si mo n di The continuity between the
endoplasmic reticulum and the inner, osmiophilic layer
of the pellicle of macrogametocytes of me 1eagridis,
indicates that this material may also play some role in the
changes in pellicular structure that occur prior to
their release from the host cell. Other workers have
suggested that the osmiophilic bodies may have a similar
function and aid in dissolving the host cell membrane
during gametogenesis (Aikawa et al., 1969; Rudzinska and
Trager, 1968).
Gametogenesis
Early changes in the fine structure of gametocytes of
H me Iearidis during the initial stages of gametocyte
maturation were identical to changes observed in
maturing gametocytes of co 1 umbae metchn i kov i and
ve 1ans (Bradbury and Trager, 1968a; 3radbury and Trager,
1968b; Sterling, 1972; Aikawa and Sterling, 1974a; Desser,
1972a). In all species studied to date, the gametocytes
round-up within their host cells. Soon afterward, the
outer layer of the pellicle detaches from the outer surface


Figure 80. Extracellular exf1 age 11 at in^ microgametocyte.
Mi crogametes (large arrows) that contain a
single axoneme bud from the outer surface of
the mi ergame tocyte, between disruptions (small
arrows) in the inner layer of the pellicle.
The mi ergametocyte nucleus (N) is stretched to
the base of the budding microgametes. X
58,000.
Figure 81. Extracellular exf1 age 11 at ing mi ergametocyte.
The microgametocyte nucleus is stretched to the
base of the budding microgamete (arrow). X
49,200.
Figure 82. Cross sections of microgametes. Each
microgamete contains a single axoneme (A) and a
mass of nuclear material (N). X 78,000.


277
Ak iba, i. S. Inui and R. Ishitani. 1971. Morphology and
distribution of intracellular schizonts in chickens
experimentally infected with Akib a caulleryi.
Nat. Inst. An. Hlth. Quart. 11: 109121.
American Ornithologists' Union. 1983. Checklist of
North American Birds. 6th edition. American
Ornithologists' Union, Washington, D.C. 877 pp.
Applegate, J.E. 1970. Population changes In latent malaria
infections associated with season and
corticosterone treatment. J. Parasitol. 56: 439-443.
Aragao, H.B. 1908. Sobre o cyclo evolutivo e a transmissao
do Haemoproteus columbae. Rev. Med. (Sao Paulo). 11:
416-419.
Aragao, H.B. 1916. Pesquizas sobre o "Haemoproteus
columbae" Bras. Med. 30: 353-354, 361-362.
Atkinson, C.T., E.C. Greiner and D.J. Forrester.
1983. Experimental vectors of Haemoproteus
meleagridis from wild turkeys in Flo r i da .1. WTl Dis. 9: 366-368.
Augustine, P.C. 1982. Effect of feed and water deprivation
on organ and blood characteristics of young turkeys.
Poultry Sci 61 : 796-799.
Ayala, S.C., J.M. Ramakka, V.F. Ramakka and C.E. Varela.
1977 Haemoproteus Plasmodium and hippoboscid
ectoparas i tes of Co 1 umb i~a wi 1 d doves. Rev. Inst.
Med. Trop. Sao Paulo. 19: 411-416.
Baker, J.R. 1957. A new vector of Ha emo p roteus
columbae in England. J. Protozool. 4: 204-208.
Baker, J.R. 1963. Transmission of Haemoproteus sp. of
English wood-pigeons by Ornithomyia avicularla. J.
Protozool. 10:461-465.
Baker, J.R. 1966a. Haemoproteus pa 1umbis sp.nov. (Sporozoa,
Haemosporina ) of the Eng 1ish Wood-Pigeon Co Iumb a p.
pa 1umbus. J. Protozool. 13: 515-519.
Baker, J.R. 1966b. The host-restriction of Haemoproteus
sp. indet. of the wood-pigeon Co I u mb a pT
pa 1umbus J. Protozool. 13: 406-408.


200


101
a mixed inflammatory infiltrate composed of macrophages,
heterophils, giant cells and red blood cells (Figures
23, 28). They were frequently invaded by macrophages
and heterophils. Macrophages were often adjacent to the
outer wall of ruptured mega 1 os ch i zon t s and occasionally
were seen phagocytizing merozoites that had been liberated
(Figures 24, 25). Giant cells commonly were found adjacent
to intact and degenerating cysts (Figures 23, 25). Muscle
fibers adjacent to the cysts were swollen, rounded and
hyaline and often contained small, gray to dark blue
granules which were oriented occasionally into parallel
lines (Figures 26, 27). Dark blue calcium deposits,
visible with hematoxylin and eosin as well as with von
Kossa's calcium stain, occupied much of the cytoplasm of
muscle fibers adjacent to more degenerate cysts (Figures
26, 27).
Capillaries and venules adjacent to or near
degenerating mega 1oschizonts were often occluded partially
or completely by thrombi composed of pink staining,
fibrinous material (Figure 29). Giant cells often
surrounded thrombi or were adjacent to occluded vessels.
Liver and spleen sections from the 2 birds that
died spontaneously at 22 DPI, 5 days after gametocytes
first appeared in circulating red cells, had numerous
golden yellow pigment deposits in the cytoplasm of
macrophages (Figure 30). The 2 birds that died on 19


227
lower prevalence o£ sporozoites in naturally infected C.
eden i could be a reflection of the smaller Wild Turkey
population at Paynes Prairie and the lower prevalence of
available gametocytes for potential vectors.
Studies of ? 1 a sinod i urn v i vax and P^ falciparum have
shown that development of the sporogonic stages in
mosquitoes is inhibited completely below temperatures of 15
- 20o q (Russell, 1959; Wernsdorfer, 1980). The
temperature dependent development of the sporogonic stages
is believed to be the major factor limiting the
temperate distribution of P^ vivax and P^ falciparum to
areas within the 20o C summer isotherm (Wernsdorfer,
1980). Morii et al. (1965) observed a similar temperature
dependence in the development of the sporogonic stages
of caul 1 e r y i Development of sporozoites in C.
ci rcumscr ip tus and arakawae was inhibited at
temperatures below 15 20o C. They suggested that
lower environmental temperatures may be an important factor
in the lower prevalences of caul 1e r yi in Japanese
poultry during the autumn.
Average monthly temperatures at Paynes Prairie and
Fisheating Creek are similar between May and October,
but range from 3 6o C cooler at Paynes Prairie between
November and April (Figure 17). The absence of
transmission of me 1eagr i d i s during February, March
and April, 1983 at Site A, when biting activity of Ch edeni


234
mega 1 oschizonts were not found (Farmer, 1965). Since
sequential observations of the development of g a r n h ami,
H, de s s e r i and the forms described by Farmer (1965) have
not been made, the significance of cytomere development in
mega 1 oschizonts of H^ me 1eagr i d i s cannot be determined.
The progressive formation of smaller and smaller cytomeres
is most similar to that described for mega 1oschizonts of
leucocytozoon (Huff, 1942; Khan and Falls, 1970).
However, final merozoite formation in species of
Le ucocy tozoon occurs by fragmentation of the cytomere
cytoplasm rather than by bud formation (Desser, 1970).
Cytomere formation has not been confirmed in other
species of Haemoproteus. Aragao (1908) described the
segmentation of mu 11inuc1eated schizonts of Ih columbae in
lung tissue into numerous uninucleate masses that underwent
further nuclear division and growth. Wenyon (1926) termed
these masses cytomeres. Other studies have not observed
this process during the exoerythrocytic development of
H co1umbae (Ahmed and Mohammed, 1977; Mohammed, 1965;
Bradbury and Gallucci, 1971). U1trastructura 1 studies
by Bradbury and Gallucci (1972) revealed that
mu 11inucleated masses of cytoplasm developed as clefts and
projections from the parent schizont. Since these did not
develop from discrete, uninucleate masses and did not
always detach from the parent body, Bradbury and
Gallucci (1972) used the term "pseudo-cytomere" as used by


6
exoerythrocytic schizonts that lack cytomeres, as is true
o£ the species transmitted by hippoboscid flies.
The proposed revisions in the classification have
not found wide acceptance because the gametocytes (i.e. the
main diagnostic stage) of each group cannot be
distinguished. While the revision may be a more accurate
reflection of phylogenetic relationships, the proposed
reclassification is speculative and essentially
nonfunctional at our current level of understanding.
Until the life cycles of more species are studied, most
workers have, by groups to subgeneric status.
Epizooti o 1ogy
Since the vectors of most species of Haemoproteus
are unknown, fundamental questions concerning their
epi zooti o 1ogy host specificity and pathogenicity have
remained unanswered. Only a few studies have examined
the seasonal transmission and role that various biting
arthropods may play in the epi zooti o 1ogy of these
parasites. In a study of Haemop roteus in insular
Newfoundland, Bennett and Coombs (1975) did not find
ookinetes, oocysts or sporozoites in 101 Ornithornyia
f ringi 11 ina recovered from passerine birds infected with


293
Vanderberg, J.P. 1977. PIa s mo dium be rghei :
quantitation of sporozoites injected by mosquitoes
feeding on a rodent host. Exp. Parasitol. 42:
169-181 .
Viens, P. A. Tarzaali and M. Quevillon. 1974. Inhibition
of the inmune response to pertussis vaccine during
Plasmodium berghei infection in mice. Am. J.
Trop. ivied. Hyg. 23: 346-849.
Wallace, G.D. 1973. Sa rcocy s tis in mice inoculated with
Toxop1 asma-1ike oocysts from cat feces. Science 180:
1375-1377.
von Wasielewski, T., and G. WUlker. 1918. Die Hdmoproteus
Infection des TUrmfalken. Arch. Schiffs. Tropenhyg.
22: 1-100.
Wenyon, C.M. 1926. Protozoology. 2 vols. Bailliere, Tindall
and Cox, London. 1563 pp.
Wernsdorfer, W.H. 1980. The importance of malaria in the
world. In Malaria" vol. 1. J.P. Kreier ed. Academic
Press, New York. pp. 1-93.
White, E.M., and G.F. Bennett. 1979. Avian
haemoprot e i dae. 12. The haemoprote i ds of the grouse
family Tetraonidae. Can. J. Zool. 57: 1465-1472.
Williams, J.E. 1978. Paratyphoid infections. In "Diseases
of Poultry". M.S. Hofstad, ed. Iowa State
University Press, Ames, Iowa. pp. 117-167.
Wong, T.C., and S.S. Desser. 1976. Fine structure of oocyst
transformation and the sporozoites of Leucocytozoon
dubreuili. J. Protozool. 23: 115-126.


279
Bennett, G.F., N.O. Okia, R.G. Ashford and A.G. Campbell.
1972. Avian haemoproteidae. II. Haemoproteus
enucleator sp.n. from the kingfisher Ispidina picta
(3oddaer t). J. Parasitol. 58: 1143-1147.
Bennett, G.F., M. Whiteway and C. Woodworth-Lynas. 1982.
A host-parasite catalogue of the avian haematozoa.
Mem. U. Newfoundland Occ. Pap. Biology. 5. Dept. of
Biology, Memorial University of Newfoundland, St.
John's, Newfoundland.
Bid1 ingmeyer, W.L. 1961. Field activity of adult
Cu 1icoi des furens Ann. Ent. Soc. Amer. 54: 149-156.
B i d1 ingmeyer W.L. 1969 The use of logarithms in
analyzing trap collections. Mosq. News. 29: 635-640.
Bierer, B.W., C.L. Vickers and J.B. Thomas. 1959. A
parasitism in turkeys due to a Haemoproteus-1 ike blood
parasite. J. Am. Vet. Med. Assoc"! 1 35 : 1 81-1 82.
Blanton, F.S., and W.W. Wirth. 1979. The sandflies
( C u 1 i c o i d e s ) of Florida (Dptera:
Ceratopogonidae) Florida Dept, of Agriculture and
Consumer Services. Gainesville, Florida. 204 pp.
Bowman, J.A., C.E. Hill and R.Q. Burleson. 1979. Seasonal
movements of restocked wild turkeys in North
Carolina. Proc. Ann. Conf. S.E. Assoc. Fish &
Wildl. Agencies.' 33: 212-223.
Bradbury, P.C., and B.B. Gallucci. 1971. The fine structure
of differentiating merozoites of Haemoproteus columbae
Kruse. J. Protozool. 18: 679-683"!
Bradbury, P.C., and B.B. Gallucci. 1972. Observations
on the fine structure of the schizonts of
Haemoproteus co1umbae. J. Protozool. 17: 405-414.
Bradbury, P.C., and J.F. Roberts. 1970. Early stages
in the differentiation of gametocytes of Haemoproteus
co 1 umbae Kruse. J. Protozool. 17: 405-413"!
Bradbury, P.C., and W. Trager. 1968a. The f ine structure
of the mature gametes of Haemoproteus coIumbae Kruse.
J. Protozool. 15: 89-102.
Bradbury, P.C., and W. Trager. 1968b. The fine
structure of m i crogametogenes i s in Haemop roteus
columbae Kruse. J. Protozool. 15: 700-712.


226
Glades County, at Fisheating Creek (Powell, 1967; L.E.
Williams, pers. comm.). In addition, Wild Turkeys from
northern Florida have a lower prevalence of me 1eagridis
(Forrester et al., 1974). Since many species of Cu 1 icoi des
have flight ranges of less than 1 km (Kettle, 1977),
transmission of me 1eagridis at Paynes Prairie may be
very local and dependent on the daily movements of Wild
Turkeys and their selection of roosting sites for the
night. A number of studies have shown that Wild Turkeys
maintain home ranges of from 400 1500 hectares in the
southeastern U.S. and, depending on the season, often move
in regular patterns and reuse favorite roosting sites
(Schorger, 1966; Bowman et al., 1979). Since Bennett trap
collections and light trap collections of C^ edeni were of
comparable sizes at Sites A and B, the higher
prevalences of transmission that were observed at Site
B, their earlier peaks and their longer seasonal
duration may have been due to greater use of the area by
Wild Turkeys. Several authors have reported that Wild
Turkeys prefer open woodland, pastures and old fields
for foraging and nesting (Stoddard, 1963; Lewis, 1964).
Since Site 3 was located in an ecotone between an open
field and the deciduous forest, it may have been more
attractive to Wild Turkeys and received greater use. As a
result, potential vectors of me 1 eag r i d i s would be
more likely to become infected with the parasite. The


225
individuals of
C.
hinmani
and
C. arboricola
a t
times
when natural transmission
of
H. me 1e a g ridis
wa s
h i gh
indicates that
c.
eden i is
the
primary vector
a t
Paynes
Prairie as well. However, several important differences
between the 2 study areas in northern and southern Florida
were evident. Transmission of me 1 eagr i d i s occurred at a
lower prevalence at Paynes Prairie than it did at
Fisheating Creek and was not continuous throughout the year
(Figures 13, 14, 18). It ceased during the cooler
winter months between January and April, when average
monthly temperatures were below 60o ancj wa s often
interrupted by periods of 2 or 4 weeks at each of the 2
study sites during the warmer months of the year. In
addition, Site A and Site 3 at Paynes Prairie had
significant differences in the time of onset of
transmission in the spring, the time when major and
minor peaks occurred and the time when transmission ceased
in the winter (Figures 13, 14). Unlike the collections at
Fisheating Creek, collections of specimens of O edeni were
not closely correlated with peaks in transmission of H.
me I eagridis at Paynes Prairie (Figures 13, 14).
Censuses of the /ild Turkey population in Florida over
the past 30 years have shown dramatic increases throughout
the state (Powell, 1967). Surveys have consistently shown
that the population in Alachua County, where Paynes Prairie
is located, is less than 1/10 as large as it is in


292
Sterling, C.R., and M. Aikawa. 1973. A comparative study
of gametocyte ultrastructure in avian Haemospor i d i a .
J. Protozool. 20: 81-92.
Sterling, C.R., and D.L. DeGiusti. 1972. U!trastructura1
aspects of schizogony, mature sc'nizonts and merozoites
of Haemoproteus metchnikovi. J. Parasitol. 58:
641-652.
Sterling, C.R., and D.L. DeGiusti. 1974. Fine structure of
differentiating oocysts and mature sporozoites of
Haemoproteu s metchnikovi in its intermediate host
Cli r y sops ca 1 1 i dus J. Protozool. 21: 276-283.
Stoddard, H.L. 1963. Maintenance and increase of the
eastern wild turkey on private lands of the coastal
plain of the deep southeast. Bull. Tall Timbers
Res. Station. 3:
Tanner, G.D., and E.C. Turner, Jr. 1974. Vertical
activities and host preferences of several Cu 1 icoi des
species in a southwestern Virginia forest. Mosq. News
34: 66-70.
Tarshis, I.B. 1955. Transmission of Haemoproteus 1ophortyx
O'Roke of the California quail by hippoboscid flies of
the species Sti 1bometopa impressa (Bigot) and Lynchia
hirsuta Ferris. Exp. Paras i tol T": 464-492.
Taliaferro, W.H. 1941. I mm unology of the parasitic
protozoa, in "Protozoa in Biological Research".
G.N. Calkins and F.M. Summers, eds. University
Press, New York. 830-854.
Terzakis, J.A. 1971. Transformation of the P1 asmodiuni
cynomolgi oocyst. J. Protozool. 18: 62-73.
Terzakis, J.A., H. Sprinz and R.A. Ward. 1966. Sporoblast
and sporozoite formation in P1 a smodiurn gal Iinaceum
infections of Aedes aegypt i Mi 1. Med. F3 1: 984-992.
Terzakis, J.A., H. Sprinz and R.A. Ward. 1967. The
transformation of the PI asmodiurn gal Iinaceum oocyst in
Aedes aegypti mosquitoes. J. Cell. 3io1. TT: 311-326.
Trefiak, W.D., and S.S. Desser. 1973. Crystalloid
inclusions in species of Leu cocytozoon ,
Parahaemoproteus and Plasmodium. Ti Protozool. ZD:
73-80.


Figure 66. Circulating microgametocyte. The parasite
is bound by a 3-layered pellicle with a
thickened, osmiophilic inner layer (arrow).
Other organelles include a nucleus (N),
mitochondria (M) with tubular cristae and food
vacuoles (Fv) that contain pigment. X 35,500.


163
Table 17. Classification sumnary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of mi ergametocytes.
The discriminant function derived from data
in table 15 was used to classify the
discriminant scores.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Ch u c k a r
3
(75.0)
+ 0
(0.0)
1
(25.0)
0 (0.0)
4
(100)
Pheasant
1
(25.0)
0
(0.0)
2
(50.0)
1 (25.0)
4
(100)
Turkey
1
(25.0)
0
(0.0)
3
(75.0)
0 (0.0)
4
(100)
Total
5
(41.7)
0
(0.0)
6
(50.0)
1 (8.3)
12
(100)
Priors*
0.
.3333
0.
3333
0.
.3333
+ Percent of total
* Prior probability of being assigned to that class


67


118


Figure 9. Scatter plot of capture times for specimens of
Cu 1 iciodes arborico1 a that were captured in Bennett
traps at Paynes FTairie and Fisheating Creek.
Capture time is plotted as minutes before or after
nautical sunset (reference line). Arrow indicates
mean capture time.


Page
Figure
13 Transmission vs. abundance of 3 species of
Cul ico ides at Site A at Paynes Prairie that
were able to support development of
Haemoproteus me 1eagridis 81
14 Transmission vs. abundance of 3 species of
Cu 1 i coi des at Site 3 at Paynes Prairie that
were able to support development of
Haemoproteus me 1e a g ridis 83
15 Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Paynes Prairie 85
16 Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Paynes Prairie 87
17 Average monthly temperatures at Paynes Prairie
and Fisheating Creek during 1982, 1983 and 1984 89
18 Transmission vs. abundance of 4 species of
Culicoi des at Fisheating Creek that were able
to support development of Haemoproteus
me 1 e a g r i d i s 91
19 Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Fisheating Creek 93
20 Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Fisheating Creek 95
21 Formalized portion of pectoral muscle from a
high dose bird that died spontaneously at 19
days pos t-i nf ect i on 116
22 An intact megaloschizont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection 118
23 An intact mega 1oschizont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection 118


97
The 4 birds had a number of secondary lesions
unrelated to the muscle cysts. One bird had a large white
nodule, approximately 5 mm in diameter, that occupied
the posterior portion of 1 lung. Another bird had
thickened air sacs with scattered white plaques that
were about 5 nrn in diameter. In all 4 birds, portions of
the gut were swollen and flaccid and the mucosa had
multifocal areas of discoloration. Reddish or black, tarry
mucoid material was occasionally present in the jejunum. A
bacterial culture from 1 bird was positive for
Salmone11 a enteridit i s Group 3.
A randomly selected bird in the control group was
killed and necropsied at 22 DPI to determine whether the
cysts observed in the birds that died may have resulted
from contamination with Sarcocystis. Tissue cysts were not
evident in any skeletal muscles of this bird. All
organs and tissues were grossly normal.
At 27 DPI, cloacal swabs from 2 of 3 randomly selected
low dose birds, 2 of 3 randomly selected control birds and
3 of 5 randomly selected high dose birds were positive for
Salmone1 I a enter id i t i s Group B, serotype heidelberg.
Gross observations surviving birds. At 8 weeks
post-infection, all surviving birds were necropsied. Nine
of the 12 low dose birds and 8 of the 8 high dose birds had
low numbers of fusiform, white cysts in the pectoral


250
The low dose birds were infected with the sporozoites
contained in 5 infected eden i A number of studies
of PI asinod i urn have shown that mosquitoes may inoculate only
a small percentage of their total number of salivary gland
sporozoites when they take a blood meal (Vanderberg,
1977). Similar studies have not been done with
ceratopogonids. However, since the prevalence of H.
me 1eagridis approaches 2% of the population of nulliparous
C. edeni at Fisheating Creek and since a bird may be bitten
by several thousand specimens of Cu I i coi des in a single
night, it is possible that intensities of natural infection
comparable to the low dose may be reached over a period of
several weeks. Additional research on the effects of
repeated, low level exposure of me 1eag ridis to
pen-reared Wild Turkeys would be helpful in determining the
possible significance of this parasite in the Wild
Turkey population.
Hos t Spec if i city
The present taxonomy of avian species of Haemoproteus
is based on the limited experimental evidence that suggests
that some species are specific to host family (Bennett
et al., 1982; Bennett et al., 1972). While this taxonomic
scheme provides a functional framework for dealing with the


Figure 16. Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Paynes Prairie. Normals are based on averages
of monthly totals between 1951 and 1980
(Climatological Data: Florida, 1982, 1983,
1984).


96
Pathogen icity
Experiment 1 Pathology
Gross observations spontaneous deaths. As early as
7 days post-infeetion (DPI) a number of subtle, behavioral
changes became evident among poults within the high dose
group. The birds stood with slightly drooped wings,
ruffled feathers and partially or completely closed eyes
and were less active than birds belonging to either the low
dose or control groups. Between 7 and 14 DPI, birds in the
high dose group developed a mild diarrhea which produced a
"pasty" vent. Their physical condition continued to
deteriorate and by 15 DPI, most birds exhibited lameness in
1 or both legs and severe depression. Most were emaciated,
dehydrated and anorexic.
On days 19, 20 and 22 DPI, 4 birds (33%) in the
high dose group died. The deaths occurred from 2 to 5 days
following the appearance of young gametocytes in the
peripheral circulation. At necropsy, the skeletal muscles
contained numerous fusiform, white cysts, up to 1.0 nm
in length and 0.5 mm in diameter. They were scattered
diffusely throughout the skeletal muscles, but were most
corrmon in the pectoral muscles. All were oriented parallel
to the muscle fibers. From 30 to 50% of the cysts were
discolored red by hemorrhage (Figure 21).


275
Mature merozoi tes of me 1 e a 3 rid i s are very
similar to merozoites of other haemosporidian
parasites. All have a specialized anterior end containing
3 polar rings, a pair of electron dense rhoptries and small
micronemes (Aikawa and Sterling, 1974b). Cytostomes
have been reported in merozoites of P1 a smodiurn and
Haernop roteus but have not been observed in merozoites
of species of Leucocytozoon (Aikawa and Sterling, 1974b).
They were not observed in merozoites of me 1eagridis, but
observations are limited. Other organelles, including the
nucleus, mitochondria and pellicle, were similar to
merozoites of other haemosporidian parasites (Aikawa and
Sterling, 1974b).
The large, e1ectron-1ucent vacuoles observed in mature
merozoites of me 1e a %ridis have not been reported from
other haemosporidian merozoites (Figures 90, 91). The
vacuole persists during development of the early
gametocytes and is clearly visible in Giemsa-stained blood
films. It disappears by the time gametocytes reach
maturity. The function and origin of this organelle are
unknown.


212


180
large, electron lucent, membrane-bound vacuole that
occupied from 1/4 to 1/3 of the total cytoplasmic volume
(Figures 90, 91).


194


LITERATURE CITED
Acton, H.'H., and R. Knowles. 1914. Studies on the
halteridium parasite of the pigeon, Haemoptoteus
columbas Ce 11i and San Felice. Indian J. Med. Res. TT
663-690.
Ad i e H.A. 1915. The sporogony of Haeinoproteus co 1 umbae.
Indian J. Med. Res. 2: 671-680.
Ad i e H.A. 1924. The sporogony of Haeinoproteus co 1 umbae.
3u 11. Soc. Pathol. Exot. 17: 605-613.
Adie, H.A. 1925. Nouvelle recherches sur la sporogonie de
Haeinoproteus col umbae Arch. Inst. Pasteur
Algr. 3: 9-15.
Ahmed, F.E., and A.H.H. Mohammed. 1977. Schizogony in
Maemooroteus columbae Kruse. J. Protozool. 24:
389-393.
Aik a w a M. C.G. Huff and C.P.A. Strome. 1970.
Morphological study of microgametogenesis of
Leucocytozoon simondi. J. Ultrastruct. Res. 32: 43-68 .
Aikawa, M., and C.R. Sterling. 1974a. High voltage
electron microscopy on m¡erogametogenes is of
Haemoproteus columbae. Z. Zellforsch. 147: 353-360.
Aikawa, M., and C.R. Sterling. 1974b. Intracellular
parasitic protozoa. Academic Press, New York. 76 pp.
Aikawa, M. C.G. Huff and H. Sprinz. 1969. Comparative
fine structure of the gametocytes of avian, reptilian
and mammalian malarial parasites. J. Ultrastruct.
Res. 26: 316-331.
Akey, B.L. 1S81 Mortality in Florida wild turkey poults
(Me 1e a g ri s gal lopavo o s c e o 1 a). M.S. Thesis.
University of Florida, Gainesville, Florida. 78 pp.
276


202


240
220
200
180
160
140
m 120
5
| 100
80
60
40
20
0
Culicoides edeni
X= 10.25
-140 -120 -100 -80 -60 -40 -20 0 20 40 60 80 100 120 140
CAPTURE TIME


LOG|0 (Nt I) LOG |o (Nil)


33
experimental group, incubated overnight in selenite
enrichment media and plated on MacConkey's Agar. Bacterial
colonies morphologically similar to Salmone 1 1 a spp. were
identified by biochemical reaction with Micro ID test
kits (Mallincrock Industries). Sa 1mone 1 la spp. isolates
were sent to the National Veterinary Diagnostic Laboratory
at Ames, Iowa for further typing. At the termination
of the experiment, cloacal swabs were made from all
surviving birds and screened for Sa 1mone1 la spp. as above.
At 8 weeks post-infection, all surviving birds were
killed by electrocution and necropsied. Wet weights of
heart, liver and spleen (expressed as percent of total
body weight at necropsy) were measured. Representative
pieces of pectoral muscle, liver, spleen, heart, lung,
brain, proveniriculus, gizzard, duodenum, pancreas, ileum,
jejunum, cecum, kidney and bone marrow taken from the
femur were fixed in 10% buffered formalin. The 3 lightest
birds from each group were selected and all representative
tissues from each were dehydrated in ethanol or isopropyl
alcohol, cleared in toluene for 2 hours, embedded in
paraplast, sectioned at 5 um and stained with hematoxylin
and eosin. Representative tissues from birds that died
prior to the end of the experiment were fixed in 10%
buffered formalin and Carnoy's fixative, dehydrated,
cleared, sectioned and stained as above. Selected serial
sections of skeletal muscle were stained for calcium with


268
originated from endoplasmic reticulum scattered in the
cytoplasm and from peripheral vacuolization of the oocyst.
Clefts originating from the endoplasmic reticulum were
covered on their cytoplasmic face by additional 2-layered
membranous sacs. These were lined on their innermost face
by a thin amorphous electron-dense coating analogous to the
electron, linear thickenings described under the plasma
membrane of other species.
As subdivision of the cytoplasm continues, budding
sporozoites develop under the electron dense thickenings.
Apical organelles including polar rings, micronemes and
subpe 11 icu1 ar microtubules differentiate as the sporozoites
bud from the sporoblast body. In contrast to
P 1 a s modiurn, metchnikovi does not undergo cleft
formation. Instead, budding sporozoites develop around the
periphery of the central sporoblast body (Sterling and
DeGius ti, 1974).
The differentiation of early oocysts of Leucocytozoon
occurs without vacuolization and cleft formation. As
sporozoites bud from electron dense thickenings under
the peripheral plasma membrane, the sporoblast body slowly
contracts (Desser and Wright, 1968; Desser, 1972c; Wong and
Desser, 1976; Desser and Allison, 1979). Sporozoite
differentiation and budding resembles P1 a smodiurn.
Leucocytozoon oocysts are smaller and develop fewer
sporozoites than PI asmodium oocysts. The cleft formation,


252
Tetraonidae, Me 1eagrididae and Numididae were lowered to
subfamily status and the quail were separated from the
pheasants and elevated to form the subfamily
Odontophorinae.
Prior to the major taxonomic revisions in the Order
Galliformes, results of the host specificity experiment
would have invalidated assumptions about the host
specificity of avian haemoproteids and provided evidence
that the current taxonomy of the genus Haemoproteus may be
flawed. Instead, the finding that rL me 1eagridis can be
transmitted experimentally between Turkeys (subfamily
Me 1 eagridinae), Chuckars (subfamily Phasianinae, tribe
Perdicini) and Ring-necked Pheasants (subfamily
Phasianinae, tribe Phasianini), does not invalidate the
basic assumptions of current haemoproteid taxonomy.
Unfortunately, a true test of the taxonomic scheme was not
conducted since avian hosts outside of the current
Phasianid family were not included in the experiment.
However, the conflicting interpretations, dependent on host
classification, illustrate the dangers inherent in
basing the taxonomy of 1 unrelated group of organisms on
that of another.
The results of the host specificity experiment provide
some support for the current revisions within the Order
Galliformes. Nolan et al. (1975) found that rabbit
antibodies to 9 proteins purified from domestic


281
Climatological Data: Florida. 1984. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.
Coatney, G.R. 1933. Relapse and associate phenomena in the
Haemoproteus infection of the pigeon. Am. J. Hyg. 18:
133-180.
Congdon, L.L., J.N. Farmer, B.M. Longenecker and R.P.
Bre i tcnbach. 1969. Natural and acquired
antibodies to Plasmodium 1 o p h ur a e in intact and
bursaless chickens. II. I mmunoflore scent studies.
J. Parasitol. 55: 817-824.
Cook, R.S., D.O. Trainer and W.C. Glazener. 1966.
Haemoproteus in wild turkeys from the coastal bend of
south Texas. J. Protozool. 13: 588-590.
Cowan, A.B. 1957. Reactions against mega 1osch i zonts of
Leucocytozoon simondi Mathis and Leger in ducks. J.
Infect. Dis. 100: 82-87.
Cox, F.E.G. 1978. Concomitant infections, in "Rodent
Malaria". R. Ki11 ick-Kendrick and W. Peters, eds.
Academic Press, New York. pp. 309-343.
Danilewsky, B. 1 889 La parasito 1og i e comparie du sang.
I. Nouvelles recherches sur les parasites du sang des
oiseaux. Kharkov. 93 pp.
Desser, S.S. 1967. Schizogony and gametogony of
Leucocytozoon simondi and associated reactions in the
avian nost. J. Protozool. 14: 244-254.
Desser S.S. 1970a. The fine structure of Leucocytozoon
simondi. II. Mega Ioschizoaony. Can. j. Zoo 1. 48:
417-421.
Desser, S.S. 1970b. The fine structure of Leucocytozoon
simondi. Ill. The ookinete and mature sporozoite.
Can. TT Zool. 48: 641-645.
Desser, S.S. 1972a. Gametocyte maturation, exflagellation
and fertilization in Parahaemoproteus (^Haemoproteus)
v e 1 a n s (Coatney ancl Roudabu s h ) (Haemospor i d i a :
Haemoprote i dae) an u 1 trastructura 1 study. J.
Protozool. 19: 287-296.


93
MONTHS


245
When comparisons o£ average hematocrits were made by group,
the same trends were evident in each group, indicating that
significant differences within groups may be related to the
age of the birds (Figure 39).
Deposition of pigment in macrophages of the spleen,
liver and lungs began at approximately 22 DPI, when
maturing erythrocytic gametocytes began to develop
detectable pigment granules. The clearance of parasitized
erythrocytes from the circulation was probably accomplished
by phagocytic activity of these cells, as has been
described in infections of PIasmodi um (Taliaferro, 1941).
Most pigment was deposited in the spleen, where the
major elimination of the parasite population from the
peripheral circulation occurs (Taliaferro, 1941).
Follicular hyperplasia and enlargement of the spleen,
characteristic of other species of Haemopro t eu s and
P1 asmodium, also occurred (Becker et al, 1956; Russell
et al, 1943; Taliaferro, 1941).
The significant drop in plasma protein levels in
the low dose and big!) dose birds at 1 week
post-infection may have been related to the increase in
vascular permeability that accompanies acute
inflammatory responses (Figure 40) (Smith et al.,
1972). Mahrt and Payer (1975) did not find significant
changes in total serum protein during experimental
infections of calves with Saccocys tis fusiformi s. However,


214
The total numbers of eden i and h i nina n i collected
in Bennett traps at Paynes Prairie and Fisheating Creek
were considerably larger than collections of any of the
other 10 species. While individuals of both species
were capable of supporting complete development of the
sporogonic stages of me 1 e a g r i d i s the greater
susceptibility of specimens of O edeni and their
preponderance in Bennett trap collections throughout the
year indicate that this species is the primary vector of H.
me 1e a g ridis in Florida.
Specimens of arbor icoI a made up approximately 8% of
the collections from Bennett and New Jersey traps at Paynes
Prairie. This species may be comnon enough to play a minor
role in the transmission of me 1 e a g r i d i s in northern
Florida. However, total numbers captured at Fisheating
Creek were insignificant when compared to the large numbers
of C^ eden i and C^ h i nman i that were taken at the same
time.
Cu 1 icoi des knowltoni is found rarely north of central
Florida and specimens were not captured at Paynes Prairie.
Individuals of this species composed a small fraction
(1.4%) of the Bennett Trap catch at Fisheating Creek,
but were taken more commonly in New Jersey light traps.
This discrepancy suggests that individuals of this species
use other hosts as a blood source and may not play a
significant role in the epizootiology of H_^


cause vascular congestion and edema by mechanical
interference with the circulation. In spite of the large
number of exoerythrocytic parasites, infected muscovy
ducks never developed patent infections with erythrocytic
gametocytes. Yet, inoculation of blood from white Pekin
ducks with patent ne11ionis infections reproduced the
disease in uninfected muscovy ducks. Julian and Galt
suggested that transmission occurred because inoculated
blood contained a few exoerythrocyt ic merozoites Later
work by Sibley and Werner (1984) was unable to confirm
the observations made by Julian and Galt ( 1980). They
succeeded in transmitting FU ne 11 ionis to muscovy ducks
using sporozoites from pools of naturally infected C.
downesi. They failed to observe any pathological effects
from the Haemoproteus infections. Julian et al. (1985)
have since identified the pathogenic organism as an
intracellular bacterium.
Hos t Spec ificit y
The present classification of avian haemoproteids
separates species with morphologically similar gametocytes
by host family. Since morphologically identical species
may occur in the same habitat where their hosts are exposed
to the same vectors, it is possible that many species


112
Schizonts were not detected in skeletal muscle or
other tissues in the infected bird.
Fourteen days. By 14 days post-infection, infected
birds were smaller than controls, but were otherwise
indistinguishable. Regenerated muscle fibers were
common in sections of skeletal muscle from 1 infected
bird. The regenerated areas occasionally contained
remnants of necrotic fibers that were surrounded by a
monocytic infiltrate. Perivascular, nodular infiltrates
composed of monocytes were present.
Schizonts ranged from 20 to 32 um in diameter and
developed both within and between muscle fibers (Figures
47, 48, 49). They were surrounded by a thick, hyaline wall
and were packed with dark blue, granular cytomeres that
ranged from 2 to 3 um in diameter. Schizonts were elongate
and extended as far as 90 um along the long axis of the
muscle fibers. Several schizonts were densely packed with
dark-staining granules (Figure 47). Others were surrounded
by macrophages (Figure 49).
Sections of heart from the infected bird had focal
areas of mononuclear infiltrates. Other tissues from
the infected and the control bird were unremarkable.
Seventeen days At 17 days post-infection,
pectoral muscle from the last infected bird contained a few


32
tubes were mixed with a vortex mixer and allowed to stand
for several hours. Absorbence was measured
spectrophotometrica11y at 540 nm. Hemoglobin concentration
for each paired sample was determined with a standard
measured at the same time (Cyanomethemoglobin Standard,
Fisher Scientific). Average values for each paired sample
were used in the statistical analysis.
The second hematocrit tube was spun for 5 minutes
in a microhematocrit centrifuge. The packed cell volume
(PCV) was measured and a drop of plasma was placed in
a refractometer to determine the plasma protein
concent rat ion.
Blood smears were prepared from all birds 3 times
per week as described earlier. Parasitemias were
determined by counting the number of gametocytes per 10,000
red blood cells. The number of red cells in each of 5
oil immersion fields behind the leading "tongue" of the
smear were counted. The average number of red cells per
field was determined and the number of fields needed to
scan 10,000 red cells was calculated.
At 4 weeks post-infection and again at the end of
the experiment, fecal samples were collected from each
compartment. Flotations were performed on the samples
with Sheather's sugar solution to detect coccidian
oocysts. At 4 weeks post-infection cloacal swabs were
prepared from 3-5 randomly selected birds in each


255
parasite morphology occurred. The discriminant analysis of
parasite and host cell variables was unable to correctly
classify scores from nacrogametocytes, mi ergametocytes and
host cells infected with ¡nacrogametocytes to their
respective host species. This indicates that the
garnetocytes of mel eagr i d i s and their associated changes
in host cells were essentially identical in each of the
3 host species. The discriminant analysis was more
successful in classifying host cells infected with
mi crogametocytes and correctly identified 100% of the
Ring-necked Pheasant scores. This was probably related to
the much greater lateral displacement of the host cell
nucleus that occurred in host cells of this species.
Morphometric studies of species of Leucocytozoon have
also failed to show significant morphological variation in
garnetocytes of a single species in different species of
hosts (3ennett and Campbell, 1975; Greiner and ivocan,
1977). These findings led Bennett and Campbell (1975)
to synonymize a number of mo rphologically similar
species reported from the same host family. These
synonymies were based on experimental evidence that most
species of Leucocy tozoon are specific to host family and
illustrate the importance of studies of
cross-transmission. Bennett and Campbell (1975) found that
garnetocytes of d ub r e u i 1 i L. fringillinarum and the
round garnetocytes of si mo n di could not be separated


164
Table 18. Average adjusted measurements of host cells
infected with macrogametocytes. Each variable
was divided by the average value of the same
variable from uninfected cells of the same
spec i es.
Variable
Chuckar
Pheasant
Turkey
Ce 11 Length
1.12
(0.08)+
1.10
(0.05)
1.09
(0.08)
Cell Width
1.02
(0.08)
1 .00
(0.07)
1.20
(0.08)
Ce 11 Area
1.15
(0.10)
1.18
(0.08)
1.33
(0.10)
Nucleus Length
0.94
(0.08)
0.76
(0.13)
0.90
(0.10)
Nucleus Width
0.97
(0.09)
0.96
(0.09)
0.97
(0.13)
Nucleus Area
0.95
(0.12)
0.73
(0.11)
0.86
(0.16)
NDR*
0.86
(0.28)
0.93
(0.24)
0.88
(0.16)
N =
15
7
15
+ Standard Deviation
* Nuclear Displacement Ratio (1 = no lateral displacement)
(0 = lateral displacement to the cell margin)


Figure 7 3.
Figure 74.
Maturing macrogame te The outer layer of
the pellicle has separated from the gamete and
formed membranous whorls that are lined on
their exterior by dense granules that are
the same size and electron density as free
ribosomes (arrow). X 66,700.
Maturing macrogamete. The nucleus (N) is
elongate and extends across the diameter of the
gametocyte. A nucleolus (Nu) is present at one
end of the nucleus and an electron dense
mass (large arrow) with faint, embedded
microtubules (small arrow) is located at
the opposite end. X 58,000.


270
suggested that zygotic rather than post-zygotic ineiosis had
occur red.
Attempts to trace division of the nucleus during
differentiation of PIasmodiu n oocysts have been only partly
successful. This is largely because of the difficulty
in obtaining and interpreting serial sections through
the complex, multidigitate form of the nucleus. A
number of investigators have found that nuclear division
proceeds without the disappearance of the nuclear
membrane. Electron dense masses termed centrioiar plaques
(Howells and Davies, 1971), kinetic centers (Schrevel et
al., 1977) and spindle pole bodies (Kubai, 1975) appear in
cup-like invaginations of the nuclear envelope. These are
duplicated throughout the nuclear envelope and direct a
series of multiple, asynchronous mitotic divisions.
Each kinetic center is linked with another by a thin
band of dense material. Radiating from each are spindle
microtubules of different lengths. The shorter ones appear
to be attached to electron dense kinetochores. These
are believed to be attached to individual chromosomes
(Schrevel et al., 1977). Schrevel et al. (1977) counted 8
kinetochores at each halfspindle pole, suggesting that the
haploid chromosome number in P^ her ;he i may be as few as
4. Canning and Sinden (1973) estimated that the haploid
number could be as few as 5 and possibly as high as 10 from
kinetochore counts on their micrographs of P^ berghei.


7
H. f a 1 1 i s i H. f tingi 11ae or or izivora. They found
sporozoites in 13.8% of 184 Cu 1 icoi des s ti 1obezziodes
captured in bird baited traps and suggested that it was
the sole vector. Later work by Greiner et al. (1978)
demonstrated that a number of other ornithophi 1 i c
Cu 1 icoide s were present in their study area. It is
possible that they may also contribute to transmission
of the paras i tes.
Bennett and Fa I 1 is ( 1960) found high prevalences
and high parasitemias of Haemoprot eus in resident and
migratory birds examined in June and July in Algonquin
Park, Canada, when Cu 1 icoi des populations were high.
A preponderance of low level, chronic infections occurred
during August and September when hippoboscids were
abundant. Bennett (1960) captured 6 different species
of Cu 1 icoi des with traps baited with woodland and water
birds in the same area and noted differences in
distribution of the flies by habitat. Some species, i.e.
C. s ti 1obezziodes C. sph agnumen sis, C. crepuscularis
and haematopotus, preferred the middle levels of the
forest canopy, while others, i.e. O downesi, preferred
the lake shore. Bennett (1960) also found differences
when comparisons were made by host. Cu 1 icoide s d own e s i
preferred ducks and herons to woodland birds as a blood
source. He noted that both of these characteristics may
be important in determining the types of blood parasites


162
Table 16. Cl assi £ication sumnary o£ a nearest neighbor
analysis o£ a set of calibration data composed
of adjusted measurements of mi crogametocytes.
Discriminant scores were classified with a
discriminant function derived from the
calibration data set summarized in Table 15.
Classified Into Species
Spec ies
Ch u c k a r
Pheasant
Turkey
Other
Total
Ch u c ka r
11
(73.3)+
2
(13.3)
2
(13.3)
0 (0.0)
15
(100)
Pheasant
0
(0.0)
9
(60.0)
5
(33.3)
1 (6.7)
15
(100)
Turkey
3
(20.0)
2
(13.3)
10
(66.7)
0 (0.0)
15
(100)
Tota 1
14
(31.1)
13
(28.9)
17
(37.8)
1 (2.2)
45
(100)
Priors*
0.
3333
0.
3333
0
. 3333
+ Percent of total
* Prior Probability of being assigned to that class


2
physiology of avian haemoproteids have been limited to
the few species that infect columbiformes, i.e. H.
co1umbae, H. sacha covi, H. macea 11umi and fK pa 1umbi s .
Pigeons and doves are relatively inexpensive, easy to
breed in captivity and can produce a reliable, although
limited, supply of uninfected young for experimental work.
The early discovery that at least 4 different species
of ectopar a s i tic hippoboscid flies could transmit H.
co1umbae (Sergent and Sergent, 1906; Aragao, 1908, 1916;
Gondor, 1915) led to a number of classical studies of
its life cycle during the first half of this century (Acton
and Knowles, 1914; Adie, 1915, 1925; Coatney, 1933).
Further work has been dampened by the difficulty in rearing
hippoboscids in captivity and harvesting large numbers
of sporozoites for experimental infections.
With the discovery that ceratopogonids in the genus
Cu 1 icoi des could support development of the sporogonic
stages of Haemop ro teu s ne 11ion i s from wild anatids, a
potential laboratory model using domestic ducks became
available (Falls and Wood, 1957). Unfortunately,
Cu 1 icoi des are notoriously difficult to colonize and less
available than hippoboscids for experimental work. Mi 1tgen
et al. (1981) were able to obtain sporozoites of desseri
from Blossom-Headed Parakeets, Psi11 acu 1 a roseata, from
Southeast Asia by exposing infected birds to a colony


50
45
40
35
30
25
20
Crisis
I
I
-1 J L_
2 3 4
5 6 7 8
WEEKS


Figure 18. Transmission vs. abundance of 4 species of
Cu 1 i c oid e s at Fisheating Creek that were
able to ¡Up port development of Haemop r oteus
me I eag r i d i s. (# = Bennett trap catch; 0--0 =
New Jersey light trap catch). The kite diagram
at the top of the figure indicates the % of
sentinel birds that became infected with
Ha emoproteus me 1e ag rid i s during monthly
collecting FTips between March, 1983 and
September, 1984. Sentinels were not exposed
during the October, 1984 trip. None of the
sentinels exposed during the November, 1984
collecting trip became infected. Marks on the
x-axis that follow each month mark the
middle of that month.


64
Table 9. Yearly prevalence of Haemoproteus me 1eagridis
in specimens of Cu 1 icoi des edeni at H is heati ng
Creek.
Month
Year
Poo 1 s
# Flies
Isolations
April
1983
2
98
2
May
1983
2
23
0
June
1983
1
17
0
July
1983
4
99
3
Augus t
1983
2
42
0
September
1983
2
56
2
November
1983
7
101
I
December
1983
2
81
0
January
1984
1
9
0
February
1984
4
42
1
March
1984
9
117
5
April
1934
7
75
1
May
1984
6
56
2
Total:
49
816
17
Minimum Yearly Prevalence:
2.08%


MATERIALS AND METHODS
Epizooti o 1ogy
Sentinel Study
Between 9 May, 1 982, and 15 July, 1984, groups of
3, 2-week-old, Broad-breasted white domestic turkey poults
were exposed for 2-week periods in sentinel cages placed
at 2 sites at Paynes Prairie State Preserve 2 km SSE of
Gainesville, Florida. One site (A) was located in a mixed,
deciduous forest and had 2 sentinel cages, 1 at ground
level and a second suspended from a rope hoist 7 m above
the first. The second site (B) was approximately 1 km
from the first and was located in an ecotone between the
forest and an open field. At the second site, a single
cage was placed on the ground in a small grove of oak
trees, approximately 30 m from the edge of the main
forest. All 3 sentinel cages were screened with 2 layers
of 1.3 cm hardware cloth to allow entry of vectors and
to restrict entry of large predators. Sentinel turkeys
were fed a high protein, commercial, unmedicated game
bird chow ad libitum and watered regularly throughout
each 2-week sentinel period. At the end of the 2-week
sentinel period, the birds were moved in a vector-proof
20


Figure 36. Average parasitemia for high dose birds ( O O ),
low dose birds ( r~-*k ) and control birds ( ).
The crisis occurred on the day when the peak parasitemia
was reached.


the most important vector in Florida. Conclusions were
based on the high susceptibility o Cu 1 i co i des eden i to the
parasite, its preponderance in biting collections, its
activity and vertical distribution in the forest canopy and
on isolations of Haemoproteus me Ieagr i d i s from naturally
infected specimens. Results of a concurrent sentinel study
indicated that transmission of the parasite occurred
year-round in southern Florida. The prevalence of
transmission in the more temperate climate of northern
Florida was lower, more variable and limited to between
April and December.
Sporozoite-induced experimental infections produced a
moderate to severe myositis in young domestic turkeys
and had significant effects on their growth and weight
gain. Pathological effects were associated with the
development of mega I osch i zonts in skeletal and cardiac
muscle. Mega 1oschizonts had a thick, hyaline wall, were
aseptate and were morphologically similar to those of
Haemoproteus des s eri and Ac throcys tis ga11i.
At least 2 generations of schizogony occurred.
First-generation schizonts matured between 5 and 8 days
post-infection and produced elongate merozoites. Second-
generation mega 1oschizonts developed between 8 and 17 days
post-infect i on and yielded spherical merozoites that
developed to form erythrocytic gametocytes.


220
1971; Tanner and Turner, 1974). In general, ornithophi 1ic
species were active in the forest canopy, where avian hosts
are presumably more available. Tanner and Turner (1974)
suggested that species of Cu 1 ic oid e s may occupy a
particular vertical stratum and search for a suitable host
within that stratum, regardless of whether it is a bird or
a mamna 1 .
Wild Turkeys normally roost at night in the middle
levels of the forest canopy (Schorger, 1966) and typically
fly to a suitable branch at sunset when individuals of
C. eden i C. arbor ico1 a and knowlton i are becoming
active. The significantly higher levels of transmission of
H. me 1e a a ridis that occurred to sentinel birds exposed
in the canopy at Paynes Prairie and Fisheating Creek
(Forrester, unpublished data) supports the evidence that
individuals of 1 or more of these species serve as vectors
of the parasite. Since total numbers of h i nman i
captured in Bennett traps and suction traps were
greatest before Wild Turkeys roost for the night and, since
this species was never captured in Bennett traps or suction
traps operated near the ground, it may play only a minor
role in the epi zooti o Iogy of me 1eagridis.
Transmission and Vector Abundance
Since the discovery of avian haemosporidians by
Danilewsky ( 1 889 ), few epizooti o 1ogica 1 studies have


Tab 1e Page
12Average adjusted measurements of
macrogametocytes 158
13 Classification sumnary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of macrogametocytes .. 159
14 Classification surrmary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of macrogametocytes .... 160
15 Average adjusted measurements of
microgametocytes 161
16 Classification summary of a nearest neighbor
analysis of a set of calibration data
composed of adjusted measurements of
mi ergametocytes 162
17 Classification surrmary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of microgametocytes .... 163
18 Average adjusted measurements of host cells
infected with macrogametocytes 164
19 Classification surrmary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with macrogametocy tes 165
20 Classification surrmary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with macrogametocytes 166
21 Average adjusted measurements of host cells
infected with microgametocytes 167
22 Classification surrmary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with microgametocytes 168
23 Classification sumnary of a nearest neighbor
analysis of set of test data composed of
adjusted measurements of host cells infected
with mi ergametocytes
i x
169


Figure 48. Fourteen-day-old megaloschizont from
pectoral muscle of an experimentally
infected turkey. The me ga 1 oschizont
contains dark, spherical cytomeres and is
surrounded by a thick, hyaline wall
(arrow). Hematoxylin and eosin. Bar = 10 urn.
Figure 49. Fourteen-day-old megaloschizont from
pectoral muscle of an experimentally
infected turkey. The megaloschizont
contains spherical, mu 1 t i nuc1eated cytomeres
and is surrounded by a hyaline wall (double
arrow) and a mononuclear infiltrate. A swollen
muscle fiber with disrupted, hyaline cytoplasm
is near the megaloschizont (arrow).
Hematoxylin and eosin. Bar = 10 urn.
Figure 50. Seventeen-day-old megaloschizont from pectoral
muscle of an experimentally infected
turkey. The megaloschizont is surrounded by
several layers of connective tissue (CT) and
contains vacuolated areas (V). Swollen,
pale and hyaline muscle fibers (arrows) are
adjacent to the megaloschizont. Hematoxylin
and eosin. Bar = 50 um.


Figure 10. Scatter plot of capture times for specimens of
Cu I icoi des knowltoni that were captured in Bennett traps
at Paynes Prairie and Fisheating Creek. Capture time is
plotted as minutes before or after nautical sunset
(reference line). Arrow indicates mean capture time.


128


Figure Page
89 A cytornere with budding merozoites 212
90 Mature merozoite 212
91 Mature merozoite 212
xv i


287
Kemp, R.L. 1978. Haemoproteus In "Diseases of Poultry".
M.S. Hofstad, ed. Towa State University Press, Ames,
Iowa. pp. 824-825.
Kettle, D.S. 1965. Biting ceratopogonids as vectors of
human and animal diseases. Acta Tropica. 22: 356-362.
Kettle, D.S. 1968a. The biting habits of Culicoides furens
(Poey) and barbosai Wirth and 31 an ton. Ti The 24-h
cycle, with a note on differences between collectors.
Bull. Ent. Res. 59: 21-31 .
Kettle, D.S. 1968b. The biting habits of Cu 1 icoi des furens
(Poey) and C. barbosai Wirth and Blanton. I I. Effect
of me t eo r o Tog i c a 1 conditions. Bull. Ent. Res. 59:
241-258.
Kettle, D.S. 1977. Biology and bionomics of bloodsucking
ceratopogonids. Ann. Rev. Entomol 22: 33-51 .
Khan, R.A., S.S. Desser and A.M. Fa Mis. 1969. Survival of
sporozoites of Leucocytozoon in birds for 11 days.
Can. J. Zool. 47: 347-350.
Khan, R.A., and A.M. Falls. 1969. Endogenous stages of
Parahaemoproteus f r i n;; i 1 1 a e (Labb, 1894 ) and
Leucocytozoon fr ingi 11inarurn Woodcock, 1910. Can. J.
Zool. 47:
Khan, R.A., and A.M. Falls. 1970. Life cycles of
Leucocytozoon dubreui 1 i Mathis and Leger, 1911 and
L e ucocyto z~o o~n fringillinarum Woodcock 1910
(Haemospor i d i a : Leucocytozoidae) J. Protozool.
17: 642-658.
Khan, R.A., and A.M. Fallis. 1971. A note on the sporogony
of Par ahaemop ro t eu s ve 1 an s (=Haemopro t eu s ve 1 ans
Co a tney and Roudabush) (Haerno s po r i d i a : Haemopro te i dae )
in species of Cu 1 icoi des. Can. J. Zool. 49: 420-421.
Klei, T.R. 1972. The fine structure of Haemoproteus
columbae sporozoites. J. Protozool. 19: 281-286.
Klei, T.R., and D.L. DeGiusti. 1973. U11rastruc tur a 1
changes in salivary glands of Pseudo 1ynchia
canariensis (Dptera: Hippoboscidae) infected with
sporozoites of Haemoproteus columbae. J. Invert.
Pathol. 22: 32 1 -72'ST


DISCUSSION
5pi zooti o 1ogy
Vectors
The functional vectors of any arthropod-borne
infection must be present at a density sufficient to
maintain transmission of the organism, must use the host
species as a regular source of blood meals and must be
susceptible to development of the parasite (Sates, 19+9;
Russell, 1 959). Of the 29 species of Cu I iciodes that were
captured in New Jersey light traps at Paynes Prairie and
Fisheating Creek, individuals of only 12 species took blood
meals from turkeys exposed in Bennett traps. Biting
collections
o f
7 of these 12 specie
s ,
i e
arboricola.
C.
crepuscularis, C. gutt
i p e
n n i s
haema topot u s ,
c.
hinmani, C. paraensis and
a
scan
have been made from both birds and mammals (Blanton and
Wirth, 1979). Engorged specimens of O baueri have been
collected previously from mammals and specimens of C.
ou s airani have been collected from birds (Blanton and
Wirth, 1979). Biting records for edeni, C. knowlton i
and O nanus have not been reported previously.
213


Figure 77. Extracellular exf1 age 11 at ing mi crogametocyte.
The remnants of the host, red blood cell
nucleus are adjacent to the gametocyte
(large arrow). The parasite has a large
diffuse nucleus (N) with aggregates of electron
dense material with embedded microtubules
adjacent to the nuclear envelope (small
arrows). Axonemes (A) and mitochondria (M) are
scattered in the cytoplasm. X 39,000.


Figure 21. Formalized portion of pectoral muscle from a
high dose bird that died spontaneously at 19
days post-infection. Mega 1oschizonts appear as
numerous white streaks (arrows) scattered
throughout the tissue. The dark flecks and
discolored areas are hemorrhagic cysts. Bar =
1 cm.


39
Mahalanobis distances (Kachigan, 1982). A similar function
was derived for microgametocytes using adjusted
measurements of gametocyte length and width, gametocyte
area and number of pigment granules. Measurements of
microgametocyte nuclei were not included because they
were diffuse and poorly defined in Gi emsa-stained blood
smears (Greiner and Forrester, 1980). Two additional
discriminant functions were derived for red blood cells
containing macrogametocytes and for red blood cells
containing microgametocytes. Adjusted measurements of
host cell length and width, host nucleus length and width,
host cell area, host nucleus area and nuclear displacement
ratio were used to derive the functions. The efficacy
of each of the 4 discriminant functions was tested using
adjusted measurements of a fresh sample of 3 or 4
gametocytes and host cells from each infected host species.
Fine Structure
All tissue processed for electron microscopy was
fixed in 3% (v/v) g 1 utara 1dehyde in 0.1% (w/v) sodium
cacodylate buffer with 4% (w/v) sucrose, ph 7.2. Tissue
was post-fixed in 1% (w/v) osmium tetroxide in the same
buffer, stained £n b1oc with 2% uranyl acetate in 75%
ethanol, dehydrated through a graded ethanol series,


122


114
Sections of heart from the infected bird had focal
areas of monocytic infiltrate. A single mega 1 oschizont
without an associated host response was present. Other
tissues from the infected and control birds were
un remarkab1e.
Exoerythrocytic Development in Natural Infections
In November, 1970, a male Wild Turkey was captured
near Lake Apopka, Florida. The bird died soon after
capture, was frozen and then necropsied several weeks later
by K.P.C. Na i r and D.J. Forrester (pets. comm.). At
necropsy, scattered whitish cysts, the size of millet
seeds, were noticed on the pectoral muscles. Histological
examination of the tissue revealed occasional spherical
mega 1 oschizonts from 200 400 um in diameter. The
mega 1oschizonts were surrounded by a thick hyaline wall
(Figures 57, 58). Some contained disorganized masses of
dark staining material (Figure 57). Others held
numerous spherical merozoites (Figure 58). Muscle
fibers surrounding some of the mega 1oschizonts were
pale, swollen and had hyaline cytoplasm. Some fibers
contained small, dark-staining granules (Figure 58).
Other infections with the organism were not
detected in many subsequent necropsies of Wild Turkeys from
northern and southern Florida (Forrester, pers. comm.).


58
Table 3. Engorged specimens of Cu 1 icoides captured in
Bennett traps at Paynes Prairie, May 1982 -
July 1984.
Site A*
Site
B#
Si te A
+ B
Spec ies
Total (%)
Total
(*)
Total
(%)
C. edeni
292
(42.7%)
393
(52.8%)
685
(48.0%)
C. hinmani
205
(30.0%)
169
(22.7%)
374
(26.2%)
C. scan 1 oni
21
(3.1%)
92
(12.4%)
113
(7.9%)
C. arboricola
74
(10.8%)
34
(4.6%)
108
(7.6%)
C. nanus
58
(8.5%)
15
(2.0%)
73
(5.1%)
C. baueri
12
(1.8%)
19
(2.6%)
31
(2.2%)
C. paraensis
6
(0.9%)
10
(1.3%)
16
(1.1%)
C. haematopotus
7
(1.0%)
7
(0.9%)
14
(1.0%)
C. crepuscu 1 ar i s
6
(0.9%)
3
(0.4%)
9
(0.6%)
C. guttipennis
2
(0.3%)
1
(0.1%)
3
(0.2%)
C. insignis
0
(0.0%)
1
(0.1%)
1
(0.1%)
C. ousairani
1
(0.1%)
0
(0.0%)
1
(0.1%)
Total
684
744
1,428
* Traps operated on 56 evenings for a total of 113 hours
# Traps operated on 49 evenings for a total of 98 hours


105
At the crisis, the peak parasitemia in the high
dose birds reached an average high of 5,760 gametocytes per
10,000 red cells. Two birds had peak parasitemias that
exceeded 7,000 gametocytes per 10,000 red cells. In
many birds, more than 50% of the red cells contained
developing gametocytes. Multiple infections of red
cells were common, with some cells containing as many as 6
gametocytes. At the crisis, low dose birds had an average
peak parasitemia of 2,109 gametocytes per 10,000 red cells,
less than half that of the high dose group.
Weight. Statistical analysis of the weight data
revealed that all 4 variables in the model statement,
i.e. treatment, subject(treatment), week,
treatment*week, were highly significant (p< .0001).
When comparisons were made by week, all 3 groups were
significantly different 1 week post-infection (PI) and
at the crisis at 3 weeks PI. Weights of high dose birds
were significantly lower than control and low dose birds
during all other weeks. Other differences between the
control and low dose groups were not significant (Figure
37).
When comparisons were made within groups, control and
low dose birds had significant increases in weight at each
week PI. By contrast, average weights for the high dose
group increased during each week, but not significantly at


153
A discriminant analysis was performed on adjusted
variables from a calibration data set derived from 15
macrogametocytes from the Chuckar, 15 from the turkey and 7
from the Ring-necked Pheasant. It derived a function that
correctly classified 80% of the discriminant scores from
the Chuckar and 80% of those from the turkey (Table
13). It correctly classified only 1 (14.3%) of the 7
scores from the Ring-necked Pheasant. Five of the pheasant
scores (71.4%) were incorrectly identified as turkey and 1
(14.3%) failed to meet criteria for classification in
any of the 3 categories (Table 13).
A small data set composed of adjusted measurements of
4 macrogametocytes from the turkey, 4 from the Chuckar and
3 from the Ring-necked Pheasant was analyzed to test the
validity of the derived function. The derived function
correctly classified 100% of the turkey scores, but only
25% of the Chuckar scores and 33% of the pheasant scores -
values close to what would be expected by chance alone
(Table 14). Three (75%) of the 4 Chuckar scores were
incorrectly classified as turkey (Table 14).
Microgaroetocytes Microgametocytes from each of the 3
host species were morphologically similar and encircled the
host cell nucleus (Figures 61, 63, 65). Average
adjusted values of cell area and cell length were smallest
for m i crogametocytes from the Ring-necked Pheasant


289
Markus, M.B., and J.H. Oosthuizen. 1972. Pathogenicity of
Haemoproteus columbae. Trans. R. Soc. Trop. Med.
Hyg 60 : 186-ITT
Mehlhorn, H., W. Peters and A. Haberkorn. 1980. T'ne
formation of kinetes and oocysts in P1 a smod i um
gal 1 inaceum (Haemospor idia ) and considerations on
phylogenetic relationships between Haemosporidia,
Piroplasmida and other Coccidia. Protistologica. 16:
135-154.
Miller, L.H., M. Aikawa, J. Johnson and T. Shiroishi.
1978. interaction between cytochalasin B-treated
malarial parasites and red cells: attachment and
junction formation. J. Exp. Med. 149: 172-184.
Miller, R.E., D.W. Trampel, S.S. Desser and W.J. Soever.
1983. Leucocytozoon simondi infection in European and
Ame ricam eiders F! Am. Vet. Med. Assoc. 183:
1241-1244.
Miltgen, F., I. Landau, N. Ratanaworabhan and S. Yenbutra.
1981. Parahaemoproteus desser i n.sp.; gamitogonie et
s c h i z og o n i e chez 1 1 h 6 t e n a t u r e 1 ; Ps i ttacula
ros ea t a de Thailande, et sporogonie expr¡mental e
chez Cu 1icoi des nubeculosus Ann. Parasitol.
Hum. CompF 56: 123-130.
Mohammed. A.H.H. 1965. Studies on the schizogony of
Haemoproteus columbae Kruse 1890. Proc. Egypt.
Acad Sc i 19: TTW.
Morehouse, N.F. 1945. The occurrence of Haemoproteus in
the domesticated turkey. Tran. Am. Microsc. Soc.
64: 109-111.
Mor i i T. 1972. Presence of antigens and antibodies in
the sera of chickens infected with A k i b a
caul 1eryi. Nat. Inst. Hlth. Quart. 12: 161-167^
Morii, T. S. Kitaoka and K. Akiba. 1965. Some
investigations on the sporogony of Leucocy tozoon
caul 1 er y i in laboratory-reared biting midges oT
tour Cu 1 i c oid e s species. Nat. Inst. Anim. Hlth.
Quart. 5: 109-110.
Mundy, B.L., I.K. Barker and M.D. Rickard. 1975. The
developmental cycle of a species of Sa r cocy s t i s
occurring in dogs and sheep, with observations on the
pathogenicity in the intermediate host. Z.
Parasitenk. 46: 111-123.


176
Mi crogametocyt e s contained a large, diffuse nucleus
(Figure 77). Dense aggregates of electron dense
material with embedded microtubules were occasionally
located adjacent to the nuclear envelope (Figure 77).
Axonemes in various stages of assembly were scattered in
the cytoplasm of exf1 age 11 at ing microgametocytes and often
extended from the nucleus to the outer limiting membrane of
the parasite. In cross section, they consisted of 9
peripheral doublets of microtubules that surrounded 2
central tubules (Figures 77, 78). In longitudinal section,
the central tubules had regular striations along their
length (Figure 79). As development progressed, axonemes
budded from the outer surface of the parasite, between
interrupted portions of the osmiophilic inner layer of the
pellicle (Figure 80). Portions of the mi crogametocyte
nucleus were occasionally drawn to the base of flagellar
buds (Figures 80, 81). Exf1age11 ated microgametes
contained a single axoneme and a membrane-bound nucleus
(Figure 82).
Oocys ts
Three-day-old oocysts. Three-day-old oocysts were
subspherical in shape and surrounded by a thick, amorphous
wall (Figure 83). Oocysts were located under the basement
membrane of the midgut. The 2 structures had the same


THE EPIZOOTIOLOGY AND PATHOGENICITY OF
Haemopco me ^eagH cH s Levine, 1961,
BY
CARTER TAIT ATKINSON
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1985

To Mom and Dad

ACKNOWLEDGEMENTS
I would like to express my sincere appreciation to all
the people who made this study possible. Donald J.
Forrester and Ellis C. Greiner conceived the initial
idea and recruited me as a graduate student. Lovett E.
Williams, Jr., and David Austin provided advice and
assistance in selecting the study sites at Paynes
Prairie State Preserve and at Fisheating Creek. Logistical
support from David Austin and the Florida Game and Fresh
Water Fish Commission as well as cooperation from Ben
Swendsen and Lykes Brothers Corporation made the field work
at Fisheating Creek possible. James Richardson and
Claus Buergeldt provided invaluable help with initial
necropsies and the interpretation and description of
microscopic lesions. Randolf Carter, Laura Perkins and
Debbie Schons contributed advice on the statistical
analysis. The IFAS Central Electron Microscopy Facility
provided materials and equipment for studies of the fine
structure of Ha emo proteus me Ieagridis.
Jerome Dj am deserves thanks for the hundreds of
turkeys he helped to bleed, for his innumerable trips to
Paynes Prairie to feed sentinel birds and for his patience

at sotting Cu 1icoi des. John Bogue deserves recognition for
his help with holding and bleeding birds and for his unique
sense of humor.
Finally, 1 would like to reserve a special
acknowledgement for my fellow graduate student, Lora G.
Rickard—for help with catching Cu 1 icoi des, holding turkeys
and swatting mosquitoes at Fisheating Creek and Paynes
Prairie, for all the stimulating discussions about worms
and protozoans and, most of all, for her friendship.
This research was funded in part by the National Wild
Turkey Federation and grant number 1270-G from the Florida
Game and Freshwater Fish Conmission.

TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS i i i
LIST OF TABLES ix
LIST OF FIGURES xi
ABSTRACT xvii
INTRODUCTION I
Epizooti o Iogy 6
Pathogenicity 9
Hos t Spec ificity II
Fine Structure 13
Objectives 17
MATERIALS AND METHODS 20
Ep izootiology 20
Sentinel Study 20
Vectors 21
Activity Cycles 25
Experimental Infections 27
Estimation of Prevalence 29
Pathogen icity 30
Experiment 1 - Pathology 30
Experiment 2 - Exoerythrocytic Stages 34
Host Spec ificity 36
Infection of Hosts 36
Morphometric Analysis 37
Fine Structure 39

Page
RESULTS 42
Epizootiology 42
Vectors 42
Paynes Prairie 42
Fisheating Creek 43
Experimental Infections 44
Sporogonic development 44
Experimental transmission 46
Activity Cycles 47
Bennett trap catches 47
New Jersey trap catches 49
Sentinel Study 50
Paynes Prairie 50
Fisheating Creek 51
Transmission and Vector Abundance 52
Paynes Prairie 52
Fisheating Creek 54
Estimation of Prevalence 55
Pathogenicity 96
Experiment 1 - Pathology 96
Gross observations - spontaneous deaths 96
Gross observations - surviving birds ... 97
Microscopic observations -
spontaneous deaths 100
Microscopic observations -
surviving birds 103
Paras i temi a 104
Weight 105
Tarsometatarsal length 106
Hematocr it 106
Plasma protein concentration 107
Hemoglobin 108
Experiment 2-Exoerythrocytic Development .. 109
Three days 109
Five days 109
Eight days 110
Eleven days 1 1 1
Fourteen days 112
Seventeen days 112
Exoerythrocytic Development - Natural
Infect ions 114
Hos t Spec ificity 151
Parasitemia 15!
v i

Page
Morphometric Analysis 152
Macrogametocytes 152
Mi erógametocytes 153
Host cells - macrogametocytes 154
Host cells - microgametocytes 156
Fine Structure 157
Mature Gametocytes 157
Gametogenesis 174
Oocysts 176
Three-day-old oocysts 176
Six-day-old oocysts 177
Megaloschizonts 179
DISCUSSION 213
Epizooti o 1ogy 213
Vectors 213
Sporogonic Stages and Transmission 215
Activity Cycles 217
Transmission and Vector Abundance 220
Pathogenicity 230
Exoerythrocyt ic Development 230
Pathology 242
Host Specificity 250
Fine Structure 257
Mature Gametocytes 257
Gametogenesis 259
Macrogametogenesis 261
Mi crogametogenesis 262
Oocysts 265
Differentiation of the oocyst 266
Nuclear divisions 269
Crystalloid 272
Mega 1oschizonts 273
LITERATURE CITED 276
BIOGRAPHICAL SKETCH 294
v i i

LIST OF TABLES
Tab Ie Page
1 Proven and presumed vectors of avian
haemoproteids 4
2 New Jersey light trap collections - Paynes
Prairie 57
3 Engorged specimens of Cu 1 icoi des captured in
Bennett traps at Paynes Prairie 58
4 New Jersey light trap collections - Fisheating
Creek 59
5 Engorged specimens of Cu 1 i coi des captured in
Bennett traps at Fisheating Creek 60
6 Susceptibility of wild-caught specimens of
Cu 1 icoi des to Haemoproteus me 1eagridis 61
7 Mean capture times for specimens of Cu 1 icoi des
taken in Bennett traps at Paynes Prairie and
Fisheating Creek 62
8 Yearly prevalence of Haemoproteus me 1eagrid i s
in specimens of Cu 1 icoi des edeni at Paynes
Prairie 63
9 Yearly prevalence of Haemoproteus meleagridis
in specimens of Cu 1 icoi des edeni at Fisneating
Creek 64
10 Attempted isolations of Haemoproteus
me I eagrid i s from pools of Culi coi des hinmani,
Culicoides arboricola and Cu 1 icoi des
knowl ton i 65
11 Average organ weights at necropsy 99
v i i i

Tab 1e Page
12Average adjusted measurements of
macrogametocytes 158
13 Classification summary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of macrogametocytes .. 159
14 Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of macrogametocytes .... 160
15 Average adjusted measurements of
microgametocytes 161
16 Classification summary of a nearest neighbor
analysis of a set of calibration data
composed of adjusted measurements of
mi erógametocytes 162
17 Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of microgametocytes .... 163
18 Average adjusted measurements of host cells
infected with macrogametocytes 164
19 Classification summary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with macrogametocy tes 165
20 Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with macrogametocytes 166
21 Average adjusted measurements of host cells
infected with microgametocytes 167
22 Classification summary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with microgametocytes 168
23 Classification summary of a nearest neighbor
analysis of set of test data composed of
adjusted measurements of host cells infected
with mi erógametocytes
i x
169

LIST OF FIGURES
Figure Page
1 Ookinete of Haemoproteus me 1eagridis 67
2 Developing oocysts of Haemoproteus me Ieagridis
on the midgut of a specimen of CUTicoides
eden i 67
3 A 6-day-old, degenerating oocyst of
Haemoproteus me 1eagridis from a specimen of
Cu 1icoi des edeni 67
4 A mature, 6-day-old oocyst of Haemoproteus
me 1e a °ridis from a specimen of Cu I icoides~edeni 67
5 One of the 2 salivary glands from a specimen
of Cu 1 icoi des edeni 67
6 A crushed salivary gland from a specimen of
Cu 1 icoi des edeni 67
7 Scatter plot of capture times for specimens of
Cu 1 i co i des li i nrnan i 69
8 Scatter plot of capture times for specimens of
Cu 1 icoi des edeni 71
9 Scatter plot of capture times for specimens of
Cu 1 icoi des arbor ico1 a 73
10 Scatter plot of capture times for specimens of
Cu 1 icoi des knowltoni 75
11 Modified New Jersey suction trap catches of
specimens of Cu I icoi des eden i at Fisheating
Creek 77
12 Modified New Jersey suction trap catches of
specimens of Cu 1 icoi des hinma ni at Fisheating
Creek 79
x

Page
Figure
13 Transmission vs. abundance of 3 species of
Cul ico ides at Site A at Paynes Prairie that
were able to support development of
Haemoproteus me 1eagridis 81
14 Transmission vs. abundance of 3 species of
Cu 1 i coi des at Site 3 at Paynes Prairie that
were able to support development of
Haemoproteus me 1e a g ridis 83
15 Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Paynes Prairie 85
16 Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Paynes Prairie 87
17 Average monthly temperatures at Paynes Prairie
and Fisheating Creek during 1982, 1983 and 1984 89
18 Transmission vs. abundance of 4 species of
Culicoi des at Fisheating Creek that were able
to support development of Haemoproteus
me 1 e a g r i d i s 91
19 Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Fisheating Creek 93
20 Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Fisheating Creek 95
21 Formalized portion of pectoral muscle from a
high dose bird that died spontaneously at 19
days pos t-i nf ect i on 116
22 An intact megaloschizont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection 118
23 An intact mega 1oschizont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection 118

Pa°e
Figure
24 A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days pos t-i n f ec t i on 120
25 A degenerating megaloschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infect ion 120
26 An intact mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection.
Hematoxylin and eosin stain 122
27 A serial section of the mega 1oschizont
illustrated in Figure 26. von Kossa's stain
for calcium 122
28 A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection 124
29 A venule blocked by a thrombus in pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection 124
30 A spleen section from a high dose bird that
died spontaneously at 22 days post-infection .. 126
31 A schizont in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection 126
32 Schizonts in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection 126
33 Nodular infiltrate from pectoral muscle of a
high dose bird that was killed at 8 weeks
post-infection 128
34 A thrombus, composed of fibrinous material,
adjacent to the remnants of a degenerating
megaloschizont 128
35 A mass of degenerating muscle fibers,
infiltrated with macrophages and heterophils .. 128
x i i

Figure Page
36 Average parasitemias for high dose birds, low
dose birds and control birds 130
37 Average weights of high dose birds, low dose
birds and control birds 132
38 Average tarsometatarsal lengths of high dose
birds, low dose birds and control birds 134
39 Average hematocrits for high dose birds, low
dose birds and control birds 136
40 Average plasma protein concentrations for
high dose birds, low dose birds and control
birds 138
41 Average hemoglobin concentrations for high
dose birds, low dose birds and control birds .. 140
42 Three-day-old schizont 142
43 Five-day-old schizont 142
44 Five-day-old schizont 142
45 Eight-day-old megaloschizont 142
46 Eight-day-old mega 1oschizont 142
47 Fourteen-day-old mega 1oschizont 142
48 Fourteen-day-old mega 1oschizont 144
49 Fourteen-day-old mega 1oschizont 144
50 Seventeen-day-old mega 1oschizont 144
51 Seventeen-day-old megaloschizont 146
52 Seventeen-day-old mega 1oschizont 146
53 Disrupted muscle fiber from pectoral muscle
of a turkey with a 3-day-old experimental
infection of Haemoproteus me 1eagridis 148
x i i i

Figure Page
54 Swollen, hyaline and disrupted pectoral muscle
fibers from a turkey with a 5-day-old
experimental infection of Haemoproteus
me 1 eagr i d i s 148
55 Regenerating muscle fibers from pectoral
muscle of a turkey with an 8-day-old
experimental infection of Haemoproteus
me 1 eagr i d i s 148
56 Deteriorating 17-day-old mega 1oschizont 148
57 Megaloschizonts from pectoral muscle of a
naturally infected Wild Turkey 150
58 Mega 1oschizont from pectoral muscle of a
naturally infected Wild Turkey 150
59 Parasitemias per 10,000 red blood cells for
turkeys, the Chuckar and the Ring-necked
Pheasant with experimental infections of
Haemoproteus me 1eagridi s 171
60 Macrogametocyte of Haemoproteus me 1eagridis
from an experiment ally infected turkey 777.... 173
61 Microgametocyte of Haemoproteus me 1eagridis
from an experimentally infected turkey 173
62 Macrogametocyte of Haemoproteus me 1eagridis
from an exper¡menta 11 y infected Chuckar TT.... 173
63 Microgametocyte of Haemoproteus me 1eagridis
from an experimentally infected Chuckar TT.... 173
64 Macrogametocyte of Haemoproteus me 1eagrid i s
from an experimentally infected Ring-necked
Pheasant 173
65 Microgametocyte of Haemoproteus me 1eagridis
from an exper¡mental 1y infected Ring-necked
Pheasant 173
66 Circulating microgametocyte 182
67 Circulating macrogametocyte 184
68 Higher magnification of Figure 67 184
x i v

Figure Page
69 Circulating macrogametocyte 186
70 Maturing macrogamete 188
71 Exf1 age 11 ating microgametocyte 190
72 Higher magnification of Figure 71 190
73 Maturing macrogamete 192
74 Maturing macrogamete 192
75 Extracellular maturing macrogamete 194
76 A higher magnification of a portion of
Figure 75 194
77 Extracellular exf1 age 11 at ing mi erógametocyte .. 196
78 Extracellular exf1 age 11 ating mi erógametocyte .. 198
79 Extracellular exf1 age 11 ating microgametocyte .. 198
80 Extracellular exf1 age 11 ating microgametocyte .. 200
81 Extracellular exf1 age 11 ating microgametocyte .. 200
82 Cross sections of microgametes 200
83 Three-day-old oocyst from a specimen of
Cu 1icoi des edeni 202
84 Six-day-old oocyst from a specimen of
Culicoi des edeni 204
85 Six-day-old oocyst from a specimen of
Culicoi des eden i 204
86 Six-day-old oocyst from a specimen of
Cul i co i des eden i 206
87 Six-day-old oocyst from a specimen of
Culicoi des edeni 208
88 Megaloschizont and associated phagocytic
cells from pectoral muscle of a high dose
bird that died spontaneously at 20 days
post-infection 210
xv

Figure Page
89 A cytornere with budding merozoites 212
90 Mature merozoite 212
91 Mature merozoite 212
xv i

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EP1ZOOTIOLOGY AND PATHOGENICITY OF
Haemoproteus meleasridis Levine, 1961,
FROM FLORIDA TURKEYS
By
Carter Tait Atkinson
December, 1985
Chairman: Donald J. Forrester
Cochairman: Ellis C. Greiner
Major Department: Veterinary Medicine
Avian species of the genus Haemoproteus are comnon
haemosporidian parasites found in many families of birds.
In spite of their widespread occurrence, little is known of
their vectors, epizootiology and pathogenicity.
Ha emo proteus me 1e a g rid i s occurs in Wild Turkeys
throughout the southeastern United States. Results of this
study showed that members of at least 5 species of
ceratopogonid flies in the genus Cu 1 i coide s, i.e.
Cu 1 i c o id e s e d e n i , Cu 1 icoid e s hinma ni, Cu 1 icoide s
arbor ¡cola, Cu 1 i c o i d e s k n ow1t o ni and Cu 1 i c o i de s
haematopotus, could support development of the
sporogonic stages of the parasite.
Results of a 2-year epizooti o 1ogica 1 study of
Ha ernoproteus me 1 eagr i d i s indicated that Cu 1 i co i des eden i is
xv i i

the most important vector in Florida. Conclusions were
based on the high susceptibility oí Cu 1 i co i des eden i to the
parasite, its preponderance in biting collections, its
activity and vertical distribution in the forest canopy and
on isolations of Haemoproteus me Ieagr i d i s from naturally
infected specimens. Results of a concurrent sentinel study
indicated that transmission of the parasite occurred
year-round in southern Florida. The prevalence of
transmission in the more temperate climate of northern
Florida was lower, more variable and limited to between
April and December.
Sporozoite-induced experimental infections produced a
moderate to severe myositis in young domestic turkeys
and had significant effects on their growth and weight
gain. Pathological effects were associated with the
development of mega I osch i zonts in skeletal and cardiac
muscle. Mega 1oschizonts had a thick, hyaline wall, were
aseptate and were morphologically similar to those of
Haemoproteus des s eri and Ac throcys tis ga11i.
At least 2 generations of schizogony occurred.
First-generation schizonts matured between 5 and 8 days
post-infection and produced elongate merozoites. Second-
generation mega 1oschizonts developed between 8 and 17 days
post-infect i on and yielded spherical merozoites that
developed to form erythrocytic gametocytes.

Haemoproteus me 1e a g r i d i s was transmitted
experimentally to a Chuckar and a Ring-necked Pheasant, but
not to chickens, Guineafowl or Northern Bobwhites.
The fine structure of circulating and
ex f 1 age 11 a t i ng gametocytes was similar to that of other
avian haemoproteids. The fine structure and development of
oocysts was similar to oocysts of species of
Leucocytozoon . Mature megaloschizonts differed
u 11rastructura 11y from similar forms reported from species
of Leucocytozoonâ– 
x i x

INTRODUCTION
The protozoan genus Haemopcoteus is composed of over
130 different species of blood and tissue parasites of
birds and some reptiles (Levine and Campbell, 1971; Bennett
et al., 1982). Since the discovery of these organisms
by Danilewsky (1889) nearly a century ago, little beyond
the basic a 1 pha-taxonomy of this group has been studied.
This is primarily because laboratory studies of these
species are difficult to accomplish. Transmission of
avian haemoproteids by blood inoculation is rarely
successful because the parasites do not undergo asexual
schizogony in circulating erythrocytes. Attempts by a
number of workers to transmit the parasites using tissue
homogenates or transplants containing the exoerythrocyt ic
stages have occasionally succeeded, but have not been
reliable enough for practical use (Gondor, 1915; O'Roke,
1930; Coatney, 1933; Lastra and Coatney, 1950; Bierer
et al., 1959).
Another major obstacle to the study of this genus
has been the absence of a convenient laboratory model.
Most studies of the life cycle, fine structure and
1

2
physiology of avian haemoproteids have been limited to
the few species that infect columbiformes, i.e. H.
co1umbae, H. sacharovi , H. macea 11umi and fK p a 1umbis .
Pigeons and doves are relatively inexpensive, easy to
breed in captivity and can produce a reliable, although
limited, supply of uninfected young for experimental work.
The early discovery that at least 4 different species
of ectoparas i tic hippoboscid flies could transmit H.
co1umbae (Sergent and Sergent, 1906; Aragao, 1908, 1916;
Condor, 1915) led to a number of classical studies of
its life cycle during the first half of this century (Acton
and Knowles, 1914; Adie, 1915, 1925; Coatney, 1933).
Further work has been dampened by the difficulty in rearing
hippoboscids in captivity and harvesting large numbers
of sporozoites for experimental infections.
With the discovery that ceratopogonids in the genus
Cu 1 icoi des could support development of the sporogonic
stages of Haemop roteu s ne 11ionis from wild anatids, a
potential laboratory model using domestic ducks became
available (Fallís and Wood, 1957). Unfortunately,
Cu 1 icoi des are notoriously difficult to colonize and less
available than hippoboscids for experimental work. Mi 1tgen
et al. (1981) were able to obtain sporozoites of desseri
from Blossom-Headed Parakeets, Psi11 acu 1 a roseata, from
Southeast Asia by exposing infected birds to a colony

3
o£ O nubecu1osus , but neither the avian host nor the
vector are available to workers in this country.
Currently, vectors have been reported for fewer than
10% of the described species of Haemoproteus (Table 1).
Five of these, i.e. co1umb a e, H. sacharovi, H.
ma c c a 1 1 um i , H â–  lophortyx and FL pa 1 umb i s , have been
transmitted by the bite of hippoboscid flies. The
remaining 7 species are presumed to be transmitted by
ceratopogonid flies in the genus Cu 1 ico i de s . Several
workers have reported complete development of the
sporogonic stages in 8 species of Cu 1 icoi des and have
transmitted the parasite by i ntraperitonea1 or intravenous
inoculation of sporozoites into suitable, uninfected hosts
(Bennett et al., 1965; Kahn and Fa 1 I is, 1971; Miltgen
et al, 1981; Atkinson et al., 1983). Transmission by
bite has not been demonstrated for any haemoproteid in
this group.
The discovery that hippoboscid flies could transmit
Haemoproteus , their common occurrence on birds and the
morphological similarities among gametocytes of various
species, led to the early assumption that this group was
fairly homogeneous. The recent advances in understanding
of the complex life cycles of cyst forming coccidian
parasites such as Sarcocystis and Toxop1 a sma, has
demonstrated that sporozoan life cycles are far more

4
Table 1: Proven and presumed vectors of avian haemoproteids
Spec ies
Vector
Author
H. columbae
H. sacharovi
ITT macea 1 1 umi
HT lophor tyx
H. pa 1 umbi s
FT7 desser i~
H. ne11ionis
H. velans
H. canachi tes
H. f ringi 1 1ae
H. dan i 1ewskyi
H. me 1eagridis
Pseudolynchi a cañar iens i s
P. brunnea
FT capensis
Mi crolynchi a pus i 1 1 a
P. cañar iens i s
FT cañar iens i s
ST i 1 borne topa ~Tmpres sa
Lynch i a hirsuta
Ornithorny i a avicubr i a
Cu 1icoi des nubeculosus
o
c.
downe si
s ti 1obezziodes
c.
sphagnumensis
C. sphagnumensis
C. crepuscularis*
C.
s ti 1obezziodes*
C.
crepuscular i s*
C.
s ti 1obezziodes*
C.
sphagnumensis*
c
eden i
Cl
hinma ni
c.
arboricol a
Sergent and
Sergent, 1906
Aragao, 1908
Gonder, 1915
Aragao, 1916
Huff, 1932
Huff, 1932
O'Roke, 1930
Tar shis , 1955
Baker, 1966b
Mi 1tgen et a I . ,
1981
Fa I Iis and Wood,
1957
Kahn and Fa 11is,
1971
Kahn and Fal1is,
1971
Fallís and
Bennett, 1960
Fa 1 1 i s and
Bennett, 1961
Fa 1 1 i s and
Bennett, 1961
Bennett and
Fal lis, 1960
Bennett and
Fallís, 1960
Fallís and
Bennett, 1961
Atkinson et
al., 1983
Atkinson et
al., 1983
Atkinson et
al., 1983
Transmission not carried out, but able to support
development of oocysts and sporozoites.

5
diverse than was realized previously. There is no reason
to suggest that haemopro teids are any less so.
Accordingly, several authors have proposed that this genus
should be divided into 2 genera (Bennett et al., 1965).
They suggested that species transmitted by hippoboscid
flies, with large oocysts containing several hundred blunt
sporozoites, would remain in the genus Haemop r ot eu s â– 
Those species transmitted by ceratopogonids , with small
oocysts containing fewer than 100 pointed sporozoites,
would be placed in the genus Parahaemoproteus.
The size and presence or absence of septa in the
exoerythrocytic schizonts has been proposed as another
criterion for separating the 2 genera (Garnham, 1966).
One form of schizont, found in birds infected with H.
co1umbae, H. 1 ophor t yx, H. pa 1umbis and f r i ngi 11ae,
is sausage or oval shaped, lacks cytomeres, and occurs
in endothelial cells in a variety of tissues including
lung, liver, spleen, kidney, bone marrow, cecum and heart.
The other type, described in birds infected with H,
ga r nhami, is large and focal with diameters of 200 urn
or larger and contains numerous cytomeres, separated from
one another by septa (Garnham, 1966). Unfortunately,
schizont morphology has not correlated perfectly with
vector or oocyst size. This is exemplified by the
exoerythrocytic stages of fU f r i ngi 11ae (Khan and Fallís,
1969). This species, which develops in Cu 1 icoi des , has

6
exoerythrocytic schizonts that lack cytomeres, as is true
o£ the species transmitted by hippoboscid flies.
The proposed revisions in the classification have
not found wide acceptance because the gametocytes (i.e. the
main diagnostic stage) of each group cannot be
distinguished. While the revision may be a more accurate
reflection of phylogenetic relationships, the proposed
reclassification is speculative and essentially
nonfunctional at our current level of understanding.
Until the life cycles of more species are studied, most
workers have, by groups to subgeneric status.
Epizooti o 1ogy
Since the vectors of most species of Haemoproteus
are unknown, fundamental questions concerning their
epi zooti o 1ogy , host specificity and pathogenicity have
remained unanswered. Only a few studies have examined
the seasonal transmission and role that various biting
arthropods may play in the epi zooti o 1ogy of these
parasites. In a study of Haemop roteus in insular
Newfoundland, Bennett and Coombs (1975) did not find
ookinetes, oocysts or sporozoites in 101 Ornithornyia
f ringi 11 ina recovered from passerine birds infected with

7
H. f a 1 1 i s i , H. f tingi 11ae or or izivora. They found
sporozoites in 13.8% of 184 Cu 1 icoi des s ti 1obezziodes
captured in bird baited traps and suggested that it was
the sole vector. Later work by Greiner et al. (1978)
demonstrated that a number of other ornithophi 1 i c
Cu 1 ico i d e s were present in their study area. It is
possible that they may also contribute to transmission
of the paras i tes.
Bennett and Fa I 1 is ( 1960) found high prevalences
and high parasitemias of Haemoprot eus in resident and
migratory birds examined in June and July in Algonquin
Park, Canada, when Cu 1 icoi des populations were high.
A preponderance of low level, chronic infections occurred
during August and September when hippoboscids were
abundant. Bennett (1960) captured 6 different species
of Cu 1 icoi des with traps baited with woodland and water
birds in the same area and noted differences in
distribution of the flies by habitat. Some species, i.e.
C. s ti 1obezziodes , C. sph agnumen sis, C. crepuscularis
and haematopotus, preferred the middle levels of the
forest canopy, while others, i.e. O downesi, preferred
the lake shore. Bennett (1960) also found differences
when comparisons were made by host. Cu 1 icoide s d own e s i
preferred ducks and herons to woodland birds as a blood
source. He noted that both of these characteristics may
be important in determining the types of blood parasites

8
found in certain species of birds as well as the
specificity of the parasites themselves.
Anomalies in the prevalence of columbiform
haemoproteids have led a number of other authors to
question the importance of hippoboscid flies as vectors.
Huff (1932), Hanson et al. (1957) and Greiner (1975) felt
that the low prevalence of hippoboscids on columbiform
birds at times when the incidence of Haemoproteus is high
suggested the involvement of another vector. Greiner
(1975) collected haematopotus and crepuscularis
feeding on Mourning Doves, Zenaida macroura, in Nebraska
during periods of active transmission of Haemop roteusâ– 
He hypothesized that they may play important roles as
vectors of macea 1 1 umi and H^ sacharovi. By contrast,
other workers have found a close correlation between the
seasonal occurrence of H^ co1umbae and its hippoboscid
vector, Pseudo 1ynchi a cañar iensis, in populations of Rock
Doves, Co 1umba 1 ivia , in Michigan (Klei and DeGiusti,
1975). Ayala et al. (1977) found large numbers of
Microlynchia p u si 1 1 a and S ti 1 borne topa podos t y 1 a on
Columbian Eared Doves, Zenaida auriculata, and felt that
these hippoboscids could account alone for the high levels
of transmission of fU macea 11umi they observed throughout
the year. Unfortunately, critical experiments to test
whether columbiform haemoproteids can be transmitted by
species of Cu 1 icoi des have not been done.

9
Pathogen icity
Since the first detailed studies of Haemoproteus
in the early part of this century, many authors have
reported instances where this organism appeared to
debilitate the host. Acton and Knowles (1914), Ad i e
(1924), Coatney (1933) and Markus and Oosthuizen (1972)
each observed a pigeon, heavily infected with columbae,
that seemed weak, anemic and with a poor appetite.
Wasielewski and WUlker (1918) reported 6 fatal infections
of Haemoproteus in the thousand they examined and Becker
et al. (1956) attributed the enlarged, purplish livers
they found in domestic pigeon squabs to infections by
H. sacharovi . Many workers have speculated that H.
co1umbae and other avian haemoproteids must be pathogenic
to some extent because peak parasitemias may involve half
of the circulating erythrocytes (Levine, 1961; Garnham,
1966). Yet, there have been no experimental attempts
to measure pathological changes in infected birds, even
among columbiform hosts that can be infected fairly easily
in the laboratory.
Only 1 study has attempted to document the
pathogenicity of Haemoproteus in detail. O'Roke (1930)
found that California Quail, Lopho r tyx cal¡fornica,
infected with fL lophor tyx had variable amounts of pigment
deposited in their lungs, testes, spleens and livers.

10
The infected birds tended to have lighter body weights,
enlarged, blackened spleens and livers that were slightly
smaller than normal. O'Roke (1930) felt that the
parasitized blood cells had less oxygen carrying capacity
and were less elastic and more likely to rupture when
passing through small capillaries. He described 4 stages
of disease in birds infected with fU lophortyx: 1)
mild-chronic with no obvious signs of infection, 2)
mild-acute where birds were restless and had poor appetites
for 2 to 4 days, 3) moderate-chronic where birds were
anemic, weak and more susceptible to death by exposure
and exhaustion and 4) heavy-acute where birds lost weight,
refused food, were unable to fly and eventually died.
O'Roke commonly observed birds with moderate-chronic
infections in the field, but saw only 4 heavy-acute
infections in the several hundred birds he examined.
Unfortunately, O'Roke did not attempt to reproduce the
disease in experimentally infected quail or precisely
quantify parasitemias and course of infection as they
related to the stage of the disease.
More recently, Julian and Galt (1980) reported several
incidents of a pathogenic Haemoproteus infection in Muscovy
Ducks, Cairina mo s c h a t a , from Ontario. They found large
numbers of schizonts, morphologically similar to those
of other species of Haemop r o t eu s , in endothelial cells
from a variety of tissues. The schizonts appeared to

cause vascular congestion and edema by mechanical
interference with the circulation. In spite of the large
number of exoerythrocytic parasites, infected muscovy
ducks never developed patent infections with erythrocytic
gametocytes. Yet, inoculation of blood from white Pekin
ducks with patent ne11ionis infections reproduced the
disease in uninfected muscovy ducks. Julian and Galt
suggested that transmission occurred because inoculated
blood contained a few exoerythrocyt ic merozoites . Later
work by Sibley and Werner (1984) was unable to confirm
the observations made by Julian and Galt ( 1980). They
succeeded in transmitting Lh ne 11 ionis to muscovy ducks
using sporozoites from pools of naturally infected C.
downesi. They failed to observe any pathological effects
from the Haemoproteus infections. Julian et al. (1985)
have since identified the pathogenic organism as an
intracellular bacterium.
Hos t Spec ificit y
The present classification of avian haemoproteids
separates species with morphologically similar gametocytes
by host family. Since morphologically identical species
may occur in the same habitat where their hosts are exposed
to the same vectors, it is possible that many species

12
will need to be synonymized once their life cycles are
better known. Few of the experiments needed to confirm
the current classification have been done. Huff (1932),
Coatney (1933) and Baker (1957, 1966, 1968) studied the
host specificity of sacharovi and 1C macea 11 umi from
Mourning Doves, Zenaida macroura carol inensis, H. co1umbae
from Rock Doves, Co 1 umba 1 i v i a, and pal umbi s from Wood
Pigeons, Co 1umba pa 1umbis, respectively. Huff ( 1 932)
was able to transmit H^ s acha r ov i and H^ macea 11 umi to
Rock Doves by the bite of infected hippoboscid flies,
but Coatney (1933) was unable to infect Mourning Doves
with H^ columbae by either fly bite or inoculation of
sporozoites. Baker (1966b, 1968) attempted unsuccessfully
to transmit H^ pa 1umbis from Wood Pigeons to Rock Doves
by injection of sporozoites. Similar host restriction
was demonstrated by Fa 1 1 is and Bennett (1960) for H.
canachites f r om a Spruce Grouse, Canachi tes canadens i s.
They inoculated sporozoites obtained from Cu 1 icoides
sphagnume n sis into uninfected Ruffed Grouse, Bonasa
umb e11u s , domestic ducks, Anas bo s cha s , a Rock Dove,
Columba 1 i via, and a Java Sparrow, Padda oryzivora. Only
Ruffed Grouse became infected. These results must be
interpreted with care since the authors did not include
positive controls when they inoculated the Rock Dove and
Java Sparrow.

13
Miltgen et al. (1981) unsuccessfully attempted to
infect a parakeet, Me 1 opsi11acus sp., with HL desseri
from Blossom-Headed Parakeets by intraper i tonea 1
inoculation of sporozoites obtained from CU nubeculosusâ– 
Fine Structure
For the most part, u1trastructura 1 studies of the
genus Haemoproteus have been limited to the mature and
exf 1 age 11 ating gametocytes of columbaeâ–  Work by
Bradbury and Roberts (1970), Bradbury and Trager (1968a,
1968b), Gallucci (1974a, 1974b) and Sterling and Aikawa
(1973) has demonstrated that the morphology of these stages
is consistent with that of other haemosporidians.
Studies of the ookinetes of co1umbae and HU velans
have shown that they are similar to each other in structure
and organization, but with several important differences.
Gallucci (1974b) hypothesized that the ookinete is a
conservative stage in the life cycle of the parasite,
since it is produced sexually and more likely to retain
the organelles of its primitive ancestor. Hence,
differences between the ookinetes of these 2 species may
lend credence to the idea that haemoproteids transmitted
by Cu 1 i coi des belong in a separate taxonomic group.

14
Desser (1972a, 1972b) suggested that the large
crystalloid inclusions in ookinetes of ve 1ans originated
from amorphous, dense lipid inclusions in the
macrogametocyte. He felt that this material was converted
to a crystalline form by addition of a protein component
by a network of endoplasmic reticulum that surrounded
the precursor material. In contrast, crystalloid particles
in ookinetes of co1umbae first appeared between the
lamellae of the endoplasmic reticulum (Gallucci, 1974b).
Other differences between the ookinetes of these 2 species
occur in the fine structure of their anterior ends.
Gallucci (1974b) found a distinctive conoid in ookinetes
of co1umbae. Ookinetes of other haemosporidia ,
including fH ve 1ans, seem to lack this structure, although
observations among some species are limited (Gallucci,
1974b; Desser, 1972b). Other anterior organelles such
as conoidal rings, the apical pore, the canopy and
subpe 11 i cu 1ar microtubules appear to be present in both
H. co1umbae and ve 1ans, although Gallucci (1974b) and
Desser (1972b) interpret their micrographs differently.
Differences in interpretation are particularly evident
in descriptions of the polar ring and ribs in ookinetes
of these 2 species. Desser (1972b) interpreted structures
resembling the ribs described by Gallucci (1974b) in H.
co1umbae as an empty subpe 11 i cu 1 ar space and stated that
the subpe11 icu1 ar microtubules in ve 1 an s arose from

15
a dense ring anterior to the ribs rather than from a polar
ring, as is true of most sporozoans. Additional studies
of the ookinetes of ve 1ans or other haemoprote i ds
transmitted by Cu 1 icoi des are needed to determine whether
these differences are important.
Few studies of the schizonts and sporogonic stages
of avian haemoproteids have been conducted. Bradbury
and Gallucci (1971, 1972) examined the fine structure
of schizonts and differentiating merozoites of H.
co1umbae. They clarified a number of inconsistencies
and errors made by Aragao (1908 ) in his description of
schizonts from lung tissue in a Rock Dove. Bradbury and
Gallucci (1972) failed to find evidence of sexual
dimorphism in the schizonts they examined and suggested
that the differential staining noted by Aragao (1908)
was probably due to differences in stage of development.
Desser (1970a) described a conspicuous sexual dimorphism
in merozoites of Leucocy tozoon s imond i , but Bradbury and
Gallucci (1971) failed to find similar differences among
mature merozoites of columbaeâ– 
Bradbury and Gallucci (1971) also clarified the use
of the term "cytomere". Aragao (1908) and later Bray
(1960) used the term to refer to development of separate,
uninucleate masses that eventually underwent extensive
nuclear division to produce merozoites. Bradbury and
Gallucci (1971) found that the schizonts of H. coIumbae

16
underwent a number of cleavages to increase the surface
area available for merozoite budding. Since they did
not observe the detachment of separate nucleated masses
from the parent schizont, they suggested that the term
" p s eudocy t ome r e11 as described by Garnham (1951) should
be applied to descriptions of co 1 uinbae.
Based on their study of mature merozoites of H.
co1umbae and their comparisons to merozoites of other
haemosporidia, Bradbury and Gallucci (1972) felt that
Ha emoproteu s and P1 a smodium were more closely related
than Leucocytozoon. Mature merozoites of both PI asmodi um
and Haemoproteus appear structurally identical, while
those of Leucocytozoon differ in number of limiting
membranes, overall shape, absence of cytostomes and absence
of a mitochondrion - associated spherical body (Bradbury
and Gallucci, 1972).
Studies of the sporogonic stages of avian
haemoproteids have been limited to the sporozoites of
H. co 1 umbae. Klei ( 1972) and Klei and DeGiusti (1973)
failed to find significant differences between the
structure of these sporozoites and those of other
haemosporidians . The only u1trastructura 1 study of a
haemoproteid oocyst was conducted by Sterling and DeGiusti
(1974) on metchnikovi , a parasite of turtles that is
transmitted by the tabanid fly, Chrysops ca1 Iidus. The
small size of the oocysts they observed as well as the

17
formation of sporozoite buds from a single residual body
closely resembled oocyst development described by light
microscopy for avian haemoproteids transmitted by
Cu 1icoi de s (Fallís and Bennett, 1960; Kahn and Fallís,
1971).
Object i ves
Haemoproteus me 1eag rid i s was first reported in a
domestic turkey, Me 1eagris ga11opavo, in Texas by Morehouse
( 1945). Since then, this parasite has been found in wild
and domestic turkeys throughout the Nearctic range of
the host and in Venezuela (Greiner and Forrester, 1980).
Eve et al. (1972) and Eve et al. (1972) speculated that
H, me 1eag ridis might be a potential pathogen of wild and
domestic birds. Much of the evidence for this was circum¬
stantial, at best, and based on observations that high
parasitemias seemed to coincide with periods of high
mortality in wild birds.
The existence of a domesticated host that is
inexpensive and available throughout the year has made
H. me 1eagridis an attractive laboratory model for the
study of avian haemoproteids. Until recently this model
has not been feasible because the vectors of 1U me 1eagridis
have been unknown. In a survey of ectoparasites from

18
309 eastern Wild Turkeys collected in the southeastern
U.S., Kellog et al. (1969) only found hippoboscid flies
occasionally. Forrester (unpublished) has never recovered
hippoboscids from Florida Wild Turkeys. Despite the
uncommon occurrence of hippoboscid flies on Wild Turkeys,
the prevalence of Ha einop roteus infections is high.
Forrester et al. (1974) found 69% of 85 Wild Turkeys from
northern Florida and 87% of 399 Wild Turkeys from southern
Florida infected with this parasite. Other workers have
found prevalences ranging from 5% in Pennsylvania (Kozicky,
1948) to 80% in southern Texas (Cook, et al., 1966).
The rarity of hippoboscid flies, the high prevalence of
Haemop r ot eus infections and the fact that transmission
of the parasite to caged domestic turkeys readily occurs
when they are placed in a suitable habitat (Forrester,
et al., 1974) suggest that ceratopogonids are the primary
vectors of fC me 1eag rid i s in Florida. In confirmation
of this, Atkinson et al. (1983) recently demonstrated
that at least 3 species of Florida Cu 1 icoi des could support
complete development of the sporogonic stages of H,
me 1eag ridis.
This study was undertaken to
(1) determine the vectors and investigate the seasonal
transmission and ep i zooti o 1ogy of me 1eag rid i s in
Florida,

19
(2) study the pathogenicity of me 1eag rid i s in
domestic turkeys with controlled experimental infections,
(3) examine the host specificity of the parasite
and
(4) study the fine stucture of the me 1eag ridis
in the vertebrate and invertebrate hosts.

MATERIALS AND METHODS
Epizooti o 1ogy
Sentinel Study
Between 9 May, 1 982, and 15 July, 1984, groups of
3, 2-week-old, Broad-breasted white domestic turkey poults
were exposed for 2-week periods in sentinel cages placed
at 2 sites at Paynes Prairie State Preserve 2 km SSE of
Gainesville, Florida. One site (A) was located in a mixed,
deciduous forest and had 2 sentinel cages, 1 at ground
level and a second suspended from a rope hoist 7 m above
the first. The second site (B) was approximately 1 km
from the first and was located in an ecotone between the
forest and an open field. At the second site, a single
cage was placed on the ground in a small grove of oak
trees, approximately 30 m from the edge of the main
forest. All 3 sentinel cages were screened with 2 layers
of 1.3 cm hardware cloth to allow entry of vectors and
to restrict entry of large predators. Sentinel turkeys
were fed a high protein, commercial, unmedicated game
bird chow ad libitum and watered regularly throughout
each 2-week sentinel period. At the end of the 2-week
sentinel period, the birds were moved in a vector-proof
20

21
cage to Cu 1 icoides-proof turkey rooms and held for 4 weeks
to allow any infections acquired in the field to become
patent. They were replaced in the field on the same day
with unexposed, 2-week-old poults. The sentinel birds
were bled from a leg vein, 3 times a week for 4 weeks,
following their exposure in the field. Blood smears were
fixed with absolute methanol and stained with 10% Giemsa,
pH 7.2. Infections were diagnosed by scanning
approximately 10,000 red blood cells at 1000X. All turkeys
used in this study were obtained as day-old poults from
Thaxton's Turkeys (P.O. Box 127, Wa tki nsvi 11e, Georgia).
Vectors
Once every 2 weeks a Bennett trap (Bennett, 1960)
and a New Jersey light trap were operated at each site
within 50 m of the sentinel cages. The Bennett trap was
operated in the middle level of the forest canopy where
ornithoph i 1 i c species of Cu 1 icoi des are most active (Tanner
and Turner, 1974) from approximately I hour before sunset
to 1 hour after sunset. A turkey was placed into an 0.3
cubic meter welded wire cage made from 1.3 cm mesh and
hoisted on an 0.6 square meter plywood board into the
canopy by means of lightweight nylon rope and small
pulleys. Following an exposure of 10 - 20 minutes, the
bird was lowered quickly to the ground and covered with
an 0.6 cubic meter wooden frame screened with fine nylon

22
mesh (28 mesh per cm). Two domestic turkeys that were
the same age and size were hoisted alternately into the
canopy. While 1 bird was exposed, the second was left
undisturbed under the screened wooden frame for 10 minutes
to allow specimens of Cu 1 icoi des to complete their blood
meals. Engorged and unengorged individuals of each species
of Cu 1 icoides were then aspirated through a sleeve in
the top of the outer screened frame as they rested on
its interior.
Specimens of Cu 1 icoi des aspirated during each run
of the Bennett trap were placed into half-pint cardboard
cartons with screened tops and supplied with a cotton
pad moistened with 5% (w/v) sucrose. The beginning and
ending times for each run of the Bennett trap, temperature,
wind velocity estimated by the Beaufort scale (Oliver,
1973) and overall weather conditions were recorded. Since
early attempts at Bennett trapping were unsuccessful in
rain and when wind velocity exceeded Beaufort 3 (4-7
miles/hr.), all trapping throughout the course of the
study was done on calm evenings when rainfall was not
imminent. Between 1 and 9 days after capture, the
specimens of Cu 1 icoi des were identified (Blanton and Wirth,
1979) and classified according to parity by the method
of Dyce (1969). Bennett trap catches of individuals of
each species were expressed as the 1 og ] q of the number
captured, plus 1. When sampling was done on more than

23
1 night during a sentinel period, the geometric mean of
all samples was calculated (Bid1ingmeyer, 1969).
A standard New Jersey light trap (Hausherr's Machine
Works, Old Freehold Rd. , Toms River, NJ) equipped with
a 40 watt incandescent bulb, an automatic timer and a
delivery cone made from 40-mesh brass was operated at
each of the 2 sites 1 night approximately every 2 weeks.
Two to 3 kg of dry ice were placed in an insulated paper
envelope and hung next to the top of the trap to act as
a carbon dioxide attractant. The trap was operated in
the middle levels of the canopy at the same spot where
the Bennett trap had been placed. Sampling with the New
Jersey trap usually followed or preceded operation of
the Bennett trap by 1 day. The trap was powered by a
portable 500 watt gasoline generator and was started 30
minutes to 1 hour before sunset and run until dawn.
Insects were collected into a 1 pint mason jar containing
10% buffered formalin and a small amount of detergent
(Alconox, Fisher Scientific). Aliquots of each sample
were poured into a white enamel pan and diluted with
water. Specimens of Cu 1 icoides were picked from each
aliquot with a Pasteur pipette, identified, grouped by
sex and parity and counted. New Jersey light trap data
for each species were expressed as the log]g of the number
captured per trap night, plus 1. The geometric mean was

24
calculated when more than 1 sample was taken during a
sentinel period (Bid1ingmeyer, 1969).
Between December, 1982, and November, 1984, collecting
trips were made approximately once a month to Lykes
Fisheating Creek Wildlife Management Area at Palmdale,
Florida, 310 km SSE of Paynes Prairie. This area has
one of the most dense Wild Turkey populations in the state
(Powell, 1967; L.E. Williams, pers. comm.). Earlier
sentinel work by Forrester (unpublished) had shown that
H. meleagridis was transmitted year-round in this area.
A study site was selected in a live oak harrmock,
surrounded by cypress, at the edge of a creek swamp, 5
km SSE of Palmdale. During each collecting trip, a Bennett
trap was operated in the middle level of the canopy for
1-3 consecutive evenings and occasionally at night and
during the day. The New Jersey light trap, supplemented
with dry ice, was operated 1 night/trip in the middle
level of the canopy at a second hoist, 30 meters from
the first. Three to 5, 2-week-old domestic poults were
also exposed at the study site for the duration of each
trip. They were housed, as described earlier, I meter
above the ground in a sentinel cage. The birds were
transported to and from Gainesville in a screened,
vector-proof cage and bled as described earlier to diagnose
infections.

25
Meteorological data for the study sites at Paynes
Prairie and Fisheating Creek were obtained from the nearest
weather stations, operated on a continuous basis
(Climatological Data: Florida, 1982, 1983, 1984). In
northern Florida, data were obtained from the U.S. weather
station operated at the Gainesville airport, located
approximately 8 km N of the study areas at Paynes Prairie.
In southern Florida, data were obtained from the U.S.
weather station at LaBelle, approximately 24 km SVV of
the study site at Fisheating Creek.
Activity Cycles
Capture times were determined for specimens of C.
edeni, C. hinmani, C. arbor¡cola and C^ knowltoni obtained
in Bennett traps at Paynes Prairie and Fisheating Creek.
Capture times for individual specimens of Cu 1 icoi des were
calculated as the midpoint of the Bennett trap run when
they were collected. The data were plotted by species,
site, year and quarter (January-March, April-June,
Ju1y-September and October-December) as the number of
minutes before or after nautical sunset on the date of
capture. Sunset times were calculated from standard tables
(Nautical Almanac, Nautical Almanac Office, U.S. Naval
Observatory) and adjusted for the proper latitude and
longitude of each study site. Scatter plots for
individuals of each species were very similar when examined

26
by site, year and quarter. To determine whether capture
data from each site, year and quarter could be combined
by species and analyzed as a single data set, the mean
capture time, standard deviation and sample size was
computed for each species by site, year and quarter.
Because some species were not active during all quarters,
and a number of empty "cells" were present when mean
capture times were examined by site, year and quarter,
it was assumed that year had a minimal effect on
variability in the data. Data for each species from each
of the 3 years of the study were combined by site and
quarter. An analysis of variance using the Statistical
Analysis System, general linear models procedure was used
to test for significant interaction effects between
species*si te, spec i es*quar ter, s i te*quar ter and
spec ies*site*quarter (SAS User's Guide: Statistics, 1982).
An alpha level of 0.05 was considered significant.
Species*site and species*quarter interactions were
significant (p< 0.0001). The analysis was repeated without
the Fisheating Creek data to determine whether data from
the 2 sites at Paynes Prairie could be combined by
quarter. Species*quarter interactions were significant
(p < 0.0001). The final analysis was done by site (Site
A and B combined) and quarter with a one-way analysis
of variance using the SAS general linear models procedure
(SAS User's Guide: Statistics, 1982). Significant species

27
effects were tested with a Duncan's Multiple Range Test.
When fewer than 3 species were compared, a T-test was
used (SAS User's Guide: Basics, 1982).
A pair of New Jersey light traps, modified as
described earlier and baited with a paper envelope
containing 2-3 kg of dry ice, was operated continuously
for 24-36 hours at Fisheating Creek during the March,
April and May, 1983, collecting trips. One trap was
suspended 1 meter above the ground and the second, 7 meters
above it, in the middle level of the canopy. Both traps
were operated without light bulbs to minimize diurnal
and nocturnal differences in their attractiveness to biting
arthropods. Sample bottles from each trap were changed
every 2 hours between 1700 and 0900 hours and every 4
hours between 0900 and 1700 hours. Dry ice was replenished
every 4 hours. Data were plotted by as the average number
of individuals of each species of Cu 1 icoi des captured
per hour of trap time for each sampling period.
Experimental Infections
To test the ability of individuals of various species
of Cu 1 icoide s to support development of me 1eagrid i s ,
infected sentinel turkeys were exposed in the Bennett
trap. Engorged flies were dissected at daily intervals
up to 9 days after a blood meal was taken in 0.85 % (w/v)

28
saline or Aedes aegypti ringers (Hayes, 1953). The flies
were removed from their cartons with an aspirator and
blown into a small petri dish containing saline and a
drop of Triton 100 X (Fisher Scientific). The specimens
of Culicoides were then transferred to a glass microscope
slide and identified by wing pattern under a dissecting
scope (Blanton and Wirth, 1979). The salivary glands
and midgut were carefully removed with fine dissecting
pins (minuten nadeln) in a drop of saline, covered with
a coverslip and examined at 400X for oocysts and
sporozoites with Norma r ski contrast interference
microscopy. Questionable identifications of individual
specimens of Culicoides were confirmed after dissected
flies had been cleared overnight in liquid phenol and
mounted on a microscope slide in a drop of 50% liquid
phenol and 50% Canada balsam. Engorged midguts were
smeared on glass slide, air dried, fixed in absolute
methanol and stained with Giemsa as described earlier.
Measurements of ookinetes were made from camera lucida
drawings and adjusted to scale with a slide micrometer.
Domestic poults were infected by drawing salivary
glands containing sporozoites into the needle of a
tuberculin syringe and injecting them int r ape r i tonea 1 1 y
(IP) or intravenously (IV) into uninfected poults. Whole
flies were also ground in 0.85% saline, Aedes aegypti
ringers (Hayes, 1953) or RPMI tissue culture medium

29
containing 10% turkey serum in a glass tissue grinder
in wet ice for several minutes. The resulting slurry
was then inoculated as described into uninfected poults.
Estimates of the total number of sporozoites inoculated
were made with a hemocytometer. Permanent sporozoite
preparations were made by mixing the slurry 1:10 with
turkey serum. A drop was smeared on a glass slide, air
dried and fixed and stained as above. Measurements of
stained sporozoites were made as above.
Estimation of Prevalence
During operation of the Bennett trap, only 50-70%
of captured specimens of Cu 1 icoi des normally took a blood
meal from the bait turkey. Unengorged specimens of
Cu 1 ico i des were identified by wing pattern, grouped by
species and ground in lots of 10-20 in Aede s aegyp t i
ringers or RPMI tissue culture media, as described
earlier. The slurry from each pool was inoculated
intraperitonea11 y into separate 1- to 2-week old turkey
poults. These birds were bled as described earlier to
diagnose any infections. Minimum yearly prevalence of
H. me 1eag rid i s in pools of naturally infected eden i
was calculated for Paynes Prairie and Fisheating Creek
as the total number of positive pools / total number of
specimens of O edeni .

30
Pathogen icity
Experiment 1 - Pathology
Thirty-six, 1-day-old, female, broad-breasted white
turkey poults were obtained in October, 1984. They were
housed together in a brooder in a vector-proof room for
7 days, then banded with metal wing tags and randomly
assigned to 3 experimental groups. Birds in the first
group were inoculated IP with separate pools of 5 C edeni
that had taken blood meals from 4 domestic poults infected
with me 1eag ridis. The specimens of Cu 1 icoi des were
captured in October, 1984, at Fisheating Creek with a
Bennett trap, held for 8-9 days at room t emp erature to
allow development of sporozoites and then ground by hand
in a glass tissue grinder for several minutes in an ice
bath. The insects were triturated in RPM1 tissue culture
medium containing 10% turkey serum. The 4 domestic poults
used to infect the wild-captured specimens of eden i
had acquired their infections from a previous exposure
at Fisheating Creek. Capture of specimens of C^ edeni
occurred during 2 evenings on days 9 and 10 of patency
when most gametocytes were fully mature. All 4 birds
had similar parasitemias. Engorged C^ eden i from both
trap nights were assigned randomly to pools used to infect
the experimental poults. Sporozoites from 1 pool of
specimens of C^ eden i were counted with a hemocy tometer

31
to determine the approximate dosage. Each of the 12 birds
was inoculated with 0.5 cc of slurry containing
approximately 4,400 sporozoites.
Birds in the second experimental group were inoculated
IP with separate pools of 35 specimens of edeni , ground
and quantified as above. Each bird was inoculated with
0.5 cc of slurry containing approximately 57,500
sporozoites. Birds in the control group were inoculated
IP with 0.5 cc of RPM1 tissue culture fluid containing
10% turkey serum.
Following inoculation the birds were housed in groups
of 3 in 12 battery cages in a vector-proof room. Each
compartment held 1 bird from each experimental group.
Birds were assigned to the compartments with a random
number table. The poults were fed and watered as described
earlier.
Twenty-four hours prior to their inoculation (Week
0), and once a week for 8 weeks following infection, each
bird was weighed and the tarsometatarsal length of the
right leg was measured with calipers. Two heparinized
capillary tubes were filled with blood from a wing vein.
Blood was immediately drawn from I capillary tube into
2, 20 ul pipettes. The contents of each pipette was
iirmediately dispensed into each of 2 separate test tubes
containing 5 mis of cyanomethemag1ob i n reagent (1:251
Cyanomethemag 1 ob i n Test Kit, Fisher Scientific). The

32
tubes were mixed with a vortex mixer and allowed to stand
for several hours. Absorbence was measured
spectrophotometrica11y at 540 nm. Hemoglobin concentration
for each paired sample was determined with a standard
measured at the same time (Cyanomethemoglobin Standard,
Fisher Scientific). Average values for each paired sample
were used in the statistical analysis.
The second hematocrit tube was spun for 5 minutes
in a microhematocrit centrifuge. The packed cell volume
(PCV) was measured and a drop of plasma was placed in
a refractometer to determine the plasma protein
concent rat ion.
Blood smears were prepared from all birds 3 times
per week as described earlier. Parasitemias were
determined by counting the number of gametocytes per 10,000
red blood cells. The number of red cells in each of 5
oil immersion fields behind the leading "tongue" of the
smear were counted. The average number of red cells per
field was determined and the number of fields needed to
scan 10,000 red cells was calculated.
At 4 weeks post-infection and again at the end of
the experiment, fecal samples were collected from each
compartment. Flotations were performed on the samples
with Sheather's sugar solution to detect coccidian
oocysts. At 4 weeks post-infection cloacal swabs were
prepared from 3-5 randomly selected birds in each

33
experimental group, incubated overnight in selenite
enrichment media and plated on MacConkey's Agar. Bacterial
colonies morphologically similar to Salmone 1 1 a spp. were
identified by biochemical reaction with Micro ID test
kits (Mallincrock Industries). Sa 1mone 1 la spp. isolates
were sent to the National Veterinary Diagnostic Laboratory
at Ames, Iowa for further typing. At the termination
of the experiment, cloacal swabs were made from all
surviving birds and screened for Sa 1mone1 la spp. as above.
At 8 weeks post-infection, all surviving birds were
killed by electrocution and necropsied. Wet weights of
heart, liver and spleen (expressed as percent of total
body weight at necropsy) were measured. Representative
pieces of pectoral muscle, liver, spleen, heart, lung,
brain, proveniriculus, gizzard, duodenum, pancreas, ileum,
jejunum, cecum, kidney and bone marrow taken from the
femur were fixed in 10% buffered formalin. The 3 lightest
birds from each group were selected and all representative
tissues from each were dehydrated in ethanol or isopropyl
alcohol, cleared in toluene for 2 hours, embedded in
paraplast, sectioned at 5 um and stained with hematoxylin
and eosin. Representative tissues from birds that died
prior to the end of the experiment were fixed in 10%
buffered formalin and Carnoy's fixative, dehydrated,
cleared, sectioned and stained as above. Selected serial
sections of skeletal muscle were stained for calcium with

34
von Kossa's stain (Humason, 1979). Pectoral muscle from
2 of these birds was fixed and processed for electron
microscopy as described later.
Data on weight, tarsometatarsal length, hematocrit,
plasma protein concentration and hemoglobin were analyzed
with the SAS general linear models procedure as a
split-plot design with subjects as main plot units and
subjects at a particular time as a subplot unit (Freund
and Littel, 1983). Treatment, subject(treatment), week
and treatment*week were tested for each variable using
the Type 111 sum of squares. A p value of 0.05 or smaller
was considered significant. When treatment*weeks
interactions were significant, further comparisons were
made by treatment and by week with Duncan's Multiple Range
Test. A comparison of organ weights at necropsy was done
with a one-way analysis of variance using the SAS general
linear models procedure (SAS User's Guide: Statistics,
1982).
Experiment 2 - Exoerythrocytic Development
A second series of experimental infections was
conducted to study the development of the early
exoerythrocytic stages and their associated host response.
Four domestic turkeys, infected as sentinel birds at
Fisheating Creek, were exposed in Bennett traps at
Fisheating Creek in May, 1985, on days 8-10 of patency.

35
Engorged specimens of ed e n i and hinmani were
collected and held as described earlier for 7 to 9 days.
They were randomly assigned to pools of 48 specimens of
C. eden i and 33 specimens of h i nman i , triturated as
described earlier and inoculated IP into 6, 5-day-old,
female broad-breasted white turkey poults. One pool was
ground and quantified with a hemocytometer to estimate
total sporozoite dose. Each bird was inoculated with
1.0 cc of slurry containing approximately 169,000
sporozoites. Six control birds of the same sex and age
were inoculated intraperitonea 1 1 y with 1.0 cc of the
carrier.
On days 3, 5, 8, 11, 14 and 17 post-infect i on , 1
inoculated poult and 1 control bird were killed by
decapitation. Representative pieces of pectoral muscle,
liver, spleen, lung, heart, brain, kidney, bone marrow,
duodenum, pancreas and cecum were fixed in 10% buffered
formalin and Carnoy's fixative. Tissue fixed in Carnoy's
was dehydrated in absolute ethanol, cleared overnight
in amyl acetate and embedded in paraplast. Sections were
cut at 4 um and stained with hematoxylin and eosin or
Giemsa-colophonium (Bray and Garnham, 1962).

36
Host Speci£ icity
Infection of Hosts
In June and July, 1984, domestic turkeys infected
with H^ me 1 eagr i d i s were used as bait birds in Bennett
traps operated at Fisheating Creek. Engorged specimens
of eden i and h i nman i were collected, held at 25o
C for 7 days to allow development of sporozoites, pooled
and ground in Aedes aegyp t i Ringers (Hayes, 1953). The
slurry from the first pool of 40 specimens of edeni
and 32 specimens of hinmani, collected in June, was
divided equally among 2, 2-day-old Chukar Partridges,
A1e c t o ris c h u c k a r, 2, 2-day-old Guineafowl, Numid a
me 1eagr i s , 2, 7-day-old Ring-Necked Pheasants, Phasianus
co1chicus , and 2, 7-day-old broad-breasted white turkey
poults. The Chuckars, Guineafowl and Ring-necked Pheasants
were obtained from Morris Hatchery (Miami, Florida).
Each bird was inoculated IP with 0.15 cc of slurry.
Sporozoite counts of the slurry with a hemocytometer
revealed that each bird received approximately 375
sporozoites. Pairs of uninfected Chuckars, Guineafowl,
Ring-necked Pheasants and turkeys were kept as negative
controls. All birds were housed by species in separate
battery cages in a vector-proof room and fed and watered
as described earlier. Smears were prepared from blood
obtained from leg veins of all birds, 3 times a week,

37
for 4 weeks following inoculation. They were fixed and
stained as described earlier.
The slurry from a second pool of 37 engorged
specimens of eden i and 84 engorged specimens of C.
hinmani, collected in July, 1984, was divided equally
among 2, 3-day-old Northern Bobwhites, Colinus virginianus,
2, 7-day-old Rhode Island red chickens, Ga11u s ga11 us,
and 2, 5-day-old broad-breasted white turkeys. The
chickens were obtained from a local hatchery and the quail
were obtained from the Department of Poultry Science,
University of Florida. Each bird was inoculated IP with
0.1 cc of slurry containing approximately 5,000
sporozoites. A pair of uninfected birds of each species
was kept as negative controls. All birds were maintained
and bled as described earlier. Tissues from birds that
died prior to the end of both experiments were fixed in
10% buffered formalin and embedded, sectioned and stained
as described earlier.
Morphometric Analysis
Parasitemias were quantified as the number of
parasites per 10,000 red blood cells as described earlier.
Morphometric parameters were determined from a maximum
of 15 mature, 7- to 9-day-old mi crogametocytes and 15
mature, 7- to 9-day-old macrogametocytes from each host.
Measurements were made by the methods of Bennett and
Campbell ( 1972) as modified by Forrester et al . (1977).

38
Fifteen uninfected red cells were also measured from each
host. All measurements were made from camera lucida
drawings of infected and uninfected erythrocytes that
were adjusted to the proper scale with a slide micrometer.
A discriminant analysis was performed on measurements
of mi crogametocy t e s , mac r ogame t ocy t e s and infected host
cells from each species susceptible to fk me 1eagridis
to determine whether parasite and host cell morphology
differed in each host. Since uninfected red cells from
each host species differed in size and could have a
limiting effect on parasite size, measurements of parasites
were expressed as a percentage of the average area of
uninfected red cells from the same host. To standardize
morphological changes in host cells infected with
gametocytes, measurements from infected host cells were
expressed as a fraction of the corresponding average
measurement of uninfected host cells of the same species
(e.g. infected host cell area/average host cell area).
Before analysis, all variables were tested for normality
with the Shapiro-Wilk statistic (SAS User's Guide: Basics,
1982). Since approximately 1/4 of the variables were
non-normal, a nonparametric, nearest neighbor analysis
(k = 3) was performed (SAS User's Guide: Statistics, 1982).
Adjusted measurements of macrogametocyte length and
width, nucleus length and width, gametocyte area, nucleus
area and number of pigment granules were used as predictor
variables to generate a discriminant function based on

39
Mahalanobis distances (Kachigan, 1982). A similar function
was derived for microgametocytes using adjusted
measurements of gametocyte length and width, gametocyte
area and number of pigment granules. Measurements of
microgametocyte nuclei were not included because they
were diffuse and poorly defined in Gi emsa-stained blood
smears (Greiner and Forrester, 1980). Two additional
discriminant functions were derived for red blood cells
containing macrogametocytes and for red blood cells
containing microgametocytes. Adjusted measurements of
host cell length and width, host nucleus length and width,
host cell area, host nucleus area and nuclear displacement
ratio were used to derive the functions. The efficacy
of each of the 4 discriminant functions was tested using
adjusted measurements of a fresh sample of 3 or 4
gametocytes and host cells from each infected host species.
Fine Structure
All tissue processed for electron microscopy was
fixed in 3% (v/v) g 1 utara 1dehyde in 0.1% (w/v) sodium
cacodylate buffer with 4% (w/v) sucrose, ph 7.2. Tissue
was post-fixed in 1% (w/v) osmium tetroxide in the same
buffer, stained £n b1oc with 2% uranyl acetate in 75%
ethanol, dehydrated through a graded ethanol series,

40
cleared in acetone and embedded in Spurr's resin.
Ultrathin sections were cut on glass knives, stained with
5% (w/v) aqueous uranyl acetate and 2% (w/v) Reynold's
lead citrate and examined with a Hitachi HU-1 IE electron
microscope.
Mature gametocytes were fixed by drawing blood from
the wing vein of a turkey infected with H_^ me 1eagr i d i s
into a syringe containing the primary fixative. The
resulting clots were diced in the primary fixative, fixed
for 1 hour at room temperature, washed with 3, 10-minute
changes of buffer, post-fixed with osmium for 1 hour at
room temperature, washed with 3 more 10-minute changes
of buffer and dehydrated and embedded as described above.
Several drops of fresh blood from the same turkey were
placed on a glass slide in a humidity chamber to allow
the gametocytes to exf1 age 1 1 ate. Two minutes and 3 minutes
after the drops were made, the clots were flooded with
primary fixative and processed as above.
Specimens of edeni that had engorged on a turkey
with a heavy H^ me 1eag ridis infection, 3 and 6 days
earlier, were dissected in a drop of Aed e s aegypt i
Ringer's. The midguts were carefully removed, flooded
with a drop of primary fixative and processed through
the first series of washes. To facilitate handling and
to prevent loss during subsequent steps, the midguts were
embedded in warm 2% agar made with 0.1 M sodium cacodylate

41
buffer with 4% sucrose, pH 7.2. After they were cut into
small blocks, processing continued as above.
Pectoral muscle containing mega 1oschizonts was diced
in primary fixative and processed as described above.

RESULTS
Epizooti o 1ogy
Vectors
Paynes Prairie. During 96 nights of trapping between
May, 1982, and July, 1984, 34,342 specimens of
Cu 1icoides belonging to 27 species were captured in New
Jersey light traps at Sites A and B (Table 2). Cu 1 i co i des
insign i s was the most common species taken at Paynes
Prairie. However, 79% of the total catch of 18,996
individuals was captured in a single night at Site B in
November, 1982. Specimens of 9 additional species, i.e. C.
e d e n i ( 10.2%), stellifer (8.7%), arbor i co 1 a
(8.1%), crepuscular i s (4.1%), spinosus (3.2%), C.
s can 1 on i (2.0%), nanus (2.0%), deb i 1 i pa 1 pus (1.5%)
and n i ge r (1.4%), made up approximately 40% of the
remaining catch. Specimens of the other 17 species were
captured infrequently or in low numbers.
Of the specimens of 27 species of Cu 1 icoi des captured
in New Jersey light traps, only representatives of 12
species, totaling 1,428 engorged individuals, were taken in
42

43
turkey-baited Bennett traps at both sites. The traps were
operated on 115 different evenings for a total of 211 hours
(Table 3). Cu 1 ico i de s eden i (48.0%) and hinmani
(26.2%) were the most conrnon species and, together, made up
74.2% of the combined catch from each site. Specimens
of s can 1 oni (7.9%), arbor ico1 a (7.6%) and nanus
(5.1%) composed 20.6% of the total. Representatives of the
remaining 7 species made up 5.1% of the total Bennett trap
catch.
Fisheating Creek. During 28 nights of trapping
between December, 1982, and November, 1984, 17,857
specimens of Cu 1 ico i de s belonging to 13 species were
captured in New Jersey light traps at Fisheating Creek
(Table 4). Cu 1 icoi de s i n s i gnis (44.5%) and edeni
(40.1%) were the most cortmon species and composed 85.2% of
the catch. Cu 1 icoi des knowlton i (7.6%) and O stelIifer
(4.2%) were less common and made up 11.8% of the catch.
Representatives of the remaining 9 species were captured
infrequently or in low numbers and composed only 3% of the
total catch.
During the same period, 2,561 engorged individuals of
5 species of Cu 1 icoi des were captured in turkey-baited
Bennett traps operated on 47 different evenings for a total
of 108 hours (Table 5). Approximately 98% of the total
catch was composed of specimens of edeni (79.6%) and

44
h i runa ni (18.6%). The remaining catch was made up of
specimens of knowltoni (1.4%), arbor ico1 a (0.5%) and
C. bauer i (0.04%).
Experimental Infections
Sporogonic development. Engorged individuals of
10 species of Cu 1 icoi des were collected from Bennett traps
baited with turkeys infected with 1G me 1eagr i d i s . Fresh
preparations and Giemsa-stained smears of engorged midguts,
dissected from specimens of edeni within 24 hours after
a blood meal was taken, had numerous ookinetes. In
fresh preparations, ookinetes had a retractile "knob" or
"point" at one end and a mass of golden-brown pigment at
the other (Figure 1). Fifteen Giemsa-stained ookinetes
measured 16.5-21.1 um in length (Mean = 18.95, SD = 1.4)
and 2.5-3.75 um in width (Mean = 2.98, SD = 0.458). The
nucleus was oval to round and measured 2.0-3.25 um in
length (Mean = 2.48, SD r 0.417) and 1.5-2.25 um in
width (Mean = 1.98, SD = 0.32). In well stained
preparations, 1, and occasionally 2, empty vacuoles,
slightly smaller than the ookinete nucleus, were located
anterior and/or posterior to it.
Dissections of specimens of Cu 1 icoi des , from 2 to 7
days after they had taken blood meals from infected
turkeys, revealed both viable and degenerating oocysts

45
on the outer wall of the midguts (Figures 2, 3, 4).
Representatives of 5 species of Cu 1 icoi des, i.e. edeni,
C. arbor i col a, C. h a ema t opo t u s, C. hinmani and C.
knowltoni, were able to support complete development of H.
meleagridis and had mature oocysts, packed with
sporozoites, and salivary glands with numerous slender
sporozoites by 6 to 7 days after they had taken blood meals
(Table 6) (Figures 4, 6). Cu 1 icoi des edeni was the
most susceptible species. Almost 2/3 of the engorged
specimens of e d e n i developed salivary gland
infections (Table 6). Four oocysts from edeni were
subspherical and measured 14-16.5 um in length (Mean =
15.4, SD = 1.11) and 12-16.5 um in width (Mean = 14.1, SD =
1.84). Oocysts contained from 50-100 elongate sporozoites
that were aligned parallel to one another. A small,
eccentric residua! body composed, in part, of golden-brown
pigment granules was present in each oocyst (Figure 4).
Salivary gland sporozoites were within secretory cells
of the single major lobe that composed each of the 2 glands
(Figure 5). In fresh preparations, they often flexed
and twisted within the salivary gland. Fifteen
Gi ems a-stained sporozoites from 1 specimen of eden i
measured 9.25-12.5 um in length (Mean = 11.1, SD = 0.8) and
0.5-1.0 um in width (Mean = 0.69, SD = 0.17). The nucleus
was located approximately 1/3 of the total length from one

46
end and measured 1.0-2.0 urn in length (Mean = 1.69, SD =
0.33) and 0.5-1.0 um in width (Mean = 0.74, SD = 0.19).
Specimens of 4 species of Cu 1 i co i d e s, i.e. C.
paraensis, C. nanus, C. s can 1 on i and baue r i , were
capable of supporting partial development of me 1eagridis
and had degenerating oocysts on their outer midguts by 4 to
7 days after taking a blood meal (Table 6).
Degenerating oocysts were smaller than mature, 7-day-old
oocysts and contained large refractile granules (Figure
3). No development was observed in 3 specimens of C.
crepuscu laris (Table 6).
Experimental transmission. Salivary gland sporozoites
from specimens of C_^ e d e n i , C_^ h i nma n i , and C .
arbor i co 1 a infected 12 of 20, 1 of 6 and 1 of 2 domestic
turkeys, respectively, when inoculated IP or IV. One pool
of 5 specimens of knowltoni, ground in Aedes aegypt i
Ringer's, infected a domestic poult. A second pool of
14 specimens of knowltoni was negative, when inoculated
into another poult. Salivary gland sporozoites from a
single specimen of haematopotus did not infect a
domestic poult.
The prepatent period of all successful infections
ranged from 17-18 days, with peak red cell invasion
occurring by 18-19 days. The rate of growth and morphology
of immature and mature gametocytes was consistent with

47
descriptions of neotypes of me Ieagridis (Greiner and
Forrester, 1980).
Act ivity Cycles
Bennett trap catches. Of the 5 species of Cu 1 icoi des
capable of supporting development of me 1eagridis ,
individuals of only 4 species, O edeni, C. hinma ni , C.
a r borico1 a and knowltoni , were captured in sufficient
numbers to permit analysis of their times of capture. A
scatter plot of capture times for specimens of C.
hinmani from 2.5 hours before sunset to 2.5 hours after
sunset was unimodal with a peak at 41.6 minutes before
sundown (Figure 7). Scatter plots of capture times for
specimens of edeni, C. arboricola and O knowltoni were
also unimodal, but average peaks were 10.3, 19.2 and
28.3 minutes after sundown, respectively (Figures 8, 9,
10).
Comparisons of mean capture times among individuals of
the 4 species revealed the same trends at each site (Paynes
Prairie and Fisheating Creek) and during each quarter
(January - March, April - June, July - September, October -
December) (Table 7). Specimens of C^ hinmani had a peak in
biting activity from 24 to 59 minutes before sunset that
was significantly earlier than individuals of the other
species. Significant differences among mean capture times

48
for specimens of e d e n i , C . a r bo cicola and C.
k now 1 ton i were less clear and varied from quarter to
quarter. At Fisheating Creek, mean capture times for
specimens of arborico1 a and knowltoni were not
significantly different during any quarter. However,
the peak biting activity for specimens of edeni was
significantly earlier than specimens of arboricola
during quarters 2 and 4, but not during quarters 1 and 3.
At Paynes Prairie, specimens of edeni had a peak capture
time that was significantly earlier than specimens of C.
arboricola during quarters 3 and 4, but not during quarter
1. During quarter 2, specimens of eden i had a peak
in activity that was significantly later than specimens of
C. arbor ico1 a (Table 7).
In spite of these differences, the same trend was
evident at each site. Specimens of hinmani reached a
peak in biting activity before sunset. They were followed
from 0 to 19 minutes after sunset by specimens of C.
edeni. Specimens of arboricola were most active from 8
to 55 minutes after sunset. Specimens of knowltoni were
the last to become active between 26 and 51 minutes
after sunset. When mean capture times for individuals
of each species are compared between Paynes Prairie and
Fisheating Creek, differences are usually within 1 standard
deviation of each other.

49
New Jersey trap catches. Cu 1 icoi de s e deni and
Cu 1 i co i des h i n in a n i were the most common species
collected in the CC>2-baited New Jersey suction traps.
During the March, April and May, 1983, collecting trips to
Fisheating Creek, specimens of eden i had a peak in
activity during the 2-hour sampling period that included
sunset (Figure 11). Individuals of this species were
active at low levels during the night in March and April.
During all 3 collecting trips, activity increased following
sunrise and continued throughout the day at levels lower
than the evening peak. During the day, specimens of C.
edeni were often observed crawling on the head and on
the back feathers of sentinel turkeys exposed on the
ground. Most individuals of edeni were captured in the
suction trap that was operated in the canopy (Figure 11).
During the April and May collecting trips,
specimens of hinmani had peaks in activity during the
2-hour sampling period that included sunset and during the
early morning hours following sunrise (Figure 12).
Activity during the April trip continued into the early
afternoon. Similar diurnal activity did not occur
during the March collecting trip. All specimens of C.
hinmani were captured in the canopy trap. This species
was never captured in Bennett traps operated on the ground.

50
Sentinel Study
Paynes Prairie. Between May, 1982, and July, 1984, 30
of 327 (9.2%) sentinel poults at Site A and 32 of 140
(22.9%) sentinel poults at Site B became infected with
H. me 1 eagr i d i s â–  At Site A, 6 of 166 (3.6%) exposed on the
ground vs. 24 of 161 (14.9%) exposed in the canopy
developed patent infections. A 2 by 2, Chi Square test of
the independence of exposure height and transmission was
highly significant (p<0.01).
During 1982, transmission of me 1eagridis began
in mid-August at Site B and m i d-S ep t erab e r at Site A
(Figures 13, 14), peaked from mid-October to
mid-December during periods of above average
temperatures for northern Florida (Figure 15) and
tapered off at the end of December at Site A and in
mid-January at Site B, with the onset of cooler winter
weather in January, 1983 (Figures 15,17). As average
monthly temperatures reached and exceeded 60o F (Figure
17), transmission began again in mid-April, 1983, at Site B
and early May, 1983, at Site A and continued at both
locations throughout the summer and fall until the onset of
cooler, winter weather in mid-December, 1983 (Figures
IS, 17). Deviations from the average monthly rainfall at
Paynes Prairie were minor throughout most of the study.
Rainfall was above average during March, April, June and

51
September, 1983, and below average during July and August,
1983, and July, 1984 (Figure 16).
Two peaks in transmission occurred in 1983 at each
site. A small peak was evident between July and August and
again between November and December at Site A. Peaks at
Site B occurred 1 to 2 months earlier between May and June
and between October and November. In 1984, transmission at
both sites began in mid-April and continued until the
end of the study in July.
Fisheating Creek. Between February, 1983, and
November, 1984, 52 of 66 (78.8%) sentinel turkeys
exposed during collecting trips to Fisheating Creek became
infected with me 1eag r i d i s . Exposures as short as 24
hours in November and December, 1983, and January, March
and June, 1984, were sufficient to infect 100% of the 3 or
4 sentinel turkeys that were taken during each trip.
Between June, 1983, and September, 1984, 50 - 100% of
the sentinel birds exposed during each trip became infected
with me 1eag rid i s (Figure 18). Transmission did not
occur in February, March and April, 1983, during periods of
abnormally cool and wet weather (Figures 19, 20).
Transmission was not detected in November, 1984, when
monthly precipitation was above average and mean
temperatures were slightly below normal (Figures 19, 20).

52
Throughout the study at Fisheating Creek, average
monthly temperatures never fell below 60o F.
Transmission and Vector Abundance
Cu 1icoi de s eden i , C. hinmani, C. arboricol a and C.
knowlton i were the only species captured in sufficient
numbers in Bennett traps, to be implicated as potential
vectors at Paynes Prairie and Fisheating Creek. Because
representatives of all 4 species had distinctive peaks
in activity at sunset, biting activity was plotted for
each as the logjo 0f the number captured during the peak
Bennett trap run for a sampling night, plus 1.
Paynes Prairie. During the 2-year study at Paynes
Prairie, biting collections and light trap collections for
each species were positively correlated (Figures 13, 14).
Specimens of C^ edeni were active at variable levels
throughout the year. Bennett trap and light trap
catches were usually lowest during the cooler, winter
months between January and March (Figures 13, 14, 18).
Transmission of fC me 1eagrid i s to sentinel birds began soon
after the biting activity and abundance of C^ eden i
increased in late April, 1983 and 1984, with the onset
of warmer spring weather. Peaks in the transmission of H.
me 1 eagridis to sentinel turkeys corresponded to minor peaks

53
in the biting activity o£ eden i in July and December,
1983, at Site A (Figure 13) and in December, 1982, and May,
1983, at Site B (Figure 14). Transmission did not occur
between May and September, 1982, at either site, in
spite of relatively large, but variable, catches of C.
edeni (Figures 13, 14). Similar anomalies between the
transmission of me 1eagridis to sentinel birds and the
abundance and biting activity of eden i occurred in
October, 1982, and October, 1983, at site B (Figure 14)
when catches of edeni were low, in spite of high rates
of transmission.
Cu 1 icoi de s hinma ni was active only during the
warmer months of the year, between May and October. Both
biting activity and abundance were essentially unimodal in
distribution, with variable peaks between May and
October during periods of active transmission of H.
me 1 eag ridis to sentinel birds. Cu 1 icoi des hinman i wa s
absent during periods of high transmission of Hâ– 
me 1eagridis in December, 1982, and December, 1983 (Figures
13, 14).
Cu 1 icoi des arbor i col a was absent at both sites in 1983
and 1984 during the cooler winter months of January,
February and March. Peaks in biting activity occurred
in September, 1982, April and September, 1983, and
March, 1984, at each site. Light trap collections during
the same period at Site B had a similar, bi modal

54
distribution. A bimodal distribution was less evident
in light trap collections from Site A (Figures 13, 14).
Biting activity of a r boricol a was very low or absent
during periods of high transmission of me 1eagridis
between November and December, 1982, and between October
and December, 1983.
Fisheating Creek. Bennett trap and light trap
collections of edeni, C. h i nmani and CL arbor ico1 a from
Fisheating Creek had seasonal patterns that were similar to
catches from Paynes Prairie. Cu 1 icoides edeni remained
active throughout the year, with fewer numbers and lower
biting activity from December, 1982, through March, 1983,
and January and November, 1984, during cooler winter
weather (Figure 18). Ha emopr o t e u s me 1eag ridis was
transmitted at high levels to exposed sentinel birds
between May, 1983, and October, 1984.
Cu 1 icoi des hinmani and (L arboricola were absent or
present in low numbers from December, 1982 - March, 1983,
and from December, 1983, and March, 1984, when average
monthly temperatures were lowest (Figures 17, 18).
Light trap and Bennett trap collections of hinmani were
bimodal in distribution in 1983 with peaks in May and
September. The distribution was more unimodal in 1984.
Light trap collections of arborico1 a were variable
throughout 1983 and 1984, but were essentially unimodal

55
with several peaks between April and November of each
year. Culico ides arboricola was captured rarely in Bennett
traps and had only 1 major peak in biting activity in May,
1984 (Figure 18).
Culicoide s knowlton i was primarily a late-spring,
early-summer species, with major peaks in abundance and
biting activity between April and July, 1983, and May and
June, 1984. Fall peaks in biting activity and abundance
occurred in October, 1984.
Estimation of Prevalence
Between May, 1982, and April, 1983, unengorged
specimens of Culicoi des captured in Bennett traps at Paynes
Prairie were dissected and examined for sporozoites in the
salivary glands. Seventeen of 17 specimens of C¿ nanus, 9
of 9 C crepuscu1aris, 33 of 33 C^ arbor ico1 a and 12 of 12
C . h i n m a ni were negative for salivary gland
sporozoites. Sporozoites were found in a single,
unengorged specimen of C^ edeni, captured on 25 August,
1982, at Site A. Dissections of 209 other unengorged
specimens of C^ edeni, collected between May and April were
negative. Unfortunately, an uninfected recipient turkey
was not available for inoculation of the sporozoites to
confirm their identification.

56
Between April, 1983, and May, 1984, unengorged
specimens of Cu I icoi des captured at Paynes Prairie and
Fisheating Creek were pooled and inoculated into
domestic turkey poults. At Paynes Prairie, 2 of 9 pools of
C. eden i , totaling 343 individuals, were positive for H.
me 1eagrid i s , resulting in an estimated minimum yearly
prevalence of 0.58% (Table 8). The positive pools were
collected in November, 1983, and May, 1984, during periods
of active, natural transmission of the parasite.
At Fisheating Creek, 17 of 49 pools of C^ edeni,
totaling 816 individuals, were positive for me 1eagridis,
resulting in an estimated minimum yearly prevalence of
2.08% (Table 9). Positive pools were collected in
April, July, September and November, 1983, and February,
March and May, 1984, during active natural transmission of
the parasite.
Unengorged individuals of other species of Cu 1 icoi des
were not captured in sufficient numbers to make regular
attempts at isolation. Pools of O hinma ni, C. arboricol a
and C^ knowltoni, collected at unequal intervals throughout
the year were negative for me 1eagridis (Table 10).

57
Table 2. New Jersey light trap collections - Paynes Prairie
May 1982 - July 1984
Site A*
Site Bit
Site A +
B
Species
Total (%)
Total (%)
Total
(%)
C.
i ns ignis
1,501
(14.3%)
17,495
(73.3%)
18,996
(55.3%)
CT
eden i
2,693
(25.7%)
825
(3.5%)
3,518
(10.2%)
cr
stel1ifer
1,474
(14.1%)
1,513
(6.3%)
2,987
(8.7%)
cr
arborico 1 a
1,706
(16.3%)
1,070
(4.5%)
2,776
(8.1%)
cr
crepuscu laris
915
(8.7%)
475
(2.0%)
1,390
(4.1%)
CT
spinosus
184
(1.8%)
919
(3.9%)
1,103
(3.2%)
CT
scan 1 oni
225
(2.1%)
471
(2.0%)
696
(2.0%)
cr
nanus
606
(5.8%)
73
(0.3%)
679
(2.0%)
CT
deb i 1 i pa 1 pus
416
(4.0%)
90
(0.4%)
506
(1.5%)
CT
n i ger
120
(1.1%)
363
(1.5%)
483
(1.4%)
CT
v i 11 os ipennis
152
(1.5%)
106
(0.4%)
258
(0.8%)
CT
h i nina n i
148
(1.4%)
95
(0.4%)
243
(0.7%)
CT
paraensis
130
(1.2%)
41
(0.2%)
171
(0.5%)
CT
ousairani
61
(0.6%)
39
(0.2%)
100
(0.3%)
CT
venustus
5
(<0. 1%)
89
(0.4%)
94
(0.3%)
CT
bauer i
33
(0.3%)
37
(0.2%)
70
(0.2%)
CT
bick1eyi
25
(0.2%)
45
(0.2%)
70
(0.2%)
CT
haematopotus
40
(0.4%)
23
(0.1%)
63
(0.2%)
CT
a 1 achua
I
(< 0. 1%)
44
(0.2%)
45
(0.1%)
CT
biguttatus
19
(0.2%)
19
(0.1%)
38
(0.1%)
CT
t i s sot i
12
(0.1%)
18
(0.1%)
30
(0.1%)
CT
gut tipennis
8
(0.1%)
9
«0. 1%)
17
(0.1%)
C. pechumani
3
(< 0.1%)
0
(0.0%)
3
(< 0.1%)
CT
furens
0
(0.0%)
2
(<0. 1%)
2
(<0. 1%)
C. mulrennani
0
(0.0%)
2
(<0. 1%)
2
(<0. 1%)
CT
chiopterus
1
(< 0.1%)
0
(0.0%)
1
(<0. 1%)
CT
pi 1ifetus
0
(0.0%)
1
(<0. 1%)
1
(<0. 1%)
Total 23,864 10,478 34,342
* 52 nights of trapping
it 44 nights of trapping

58
Table 3. Engorged specimens of Cu 1 icoides captured in
Bennett traps at Paynes Prairie, May 1982 -
July 1984.
Site A*
Site
B#
Si te A
+ B
Spec ies
Total (%)
Total
(*)
Total
(%)
C. edeni
292
(42.7%)
393
(52.8%)
685
(48.0%)
C. hinmani
205
(30.0%)
169
(22.7%)
374
(26.2%)
C. scan 1 oni
21
(3.1%)
92
(12.4%)
113
(7.9%)
C. arboricola
74
(10.8%)
34
(4.6%)
108
(7.6%)
C. nanus
58
(8.5%)
15
(2.0%)
73
(5.1%)
C. baueri
12
(1.8%)
19
(2.6%)
31
(2.2%)
C. paraensis
6
(0.9%)
10
(1.3%)
16
(1.1%)
C. haematopotus
7
(1.0%)
7
(0.9%)
14
(1.0%)
C. crepuscu 1 ar i s
6
(0.9%)
3
(0.4%)
9
(0.6%)
C. guttipennis
2
(0.3%)
1
(0.1%)
3
(0.2%)
C. insignis
0
(0.0%)
1
(0.1%)
1
(0.1%)
C. ousairani
1
(0.1%)
0
(0.0%)
1
(0.1%)
Total
684
744
1,428
* Traps operated on 56 evenings for a total of 113 hours
# Traps operated on 49 evenings for a total of 98 hours

ülüIdLilüUIdlüUIdLitildl
59
Table 4. New Jersey light trap collections - Fisheating
Creek, December 1982 - November 1984. Total
catch was made on 28 different nights of
t rapping.
:c i es
Total
% Total
i ns ignis
7,940
44.5
eden i
7,279
40.8
knowltoni
1,356
7.6
s t e11ifer
742
4.2
arborico1 a
287
1.6
hinmani
129
0.7
pusi1 I us
60
0.3
ousairani
28
0.2
baueri
15
0. 1
paraensis
14
0. 1
deb i 1 i pa 1 pus
4
0.02
haematopotus
2
0.01
b i ck1eyi
1
0.01
17,857
100.00
Total

60
Table 5. Engorged specimens of Cu 1 i coi des captured in
Bennett traps at Fisheating Creek, December 1982 -
November 1984. Traps were operated on 47
different evenings for a total of 108 hours.
Spec ies
Tota 1
% Total
C. edeni
2,038
79.6
C. hinmani
475
18.6
C. knowltoni
35
1.4
C. arboricola
12
0.5
C. baueri
1
0.04
2,561 100.0
Total

61
Table 6. Susceptibility of wild-caught specimens of
Cu 1icoi des to Haemoproteus me 1eagridisâ–  Fractions
represent numbeT pos itive/ number examined and are
followed by percent of total.
Development of Parasite
Spec ies
None
Part
i a 1 *
Complete**
c
eden i
32/52
(61.5%)
c
arboricol a
6/28
(21.4%)
o
haematopotus
1/6
(16.7%)
c
hinmani
8/72
(11.1%)
c
knowltoni
1/14
(7.1%)
c
paraensis
1/2
(50.0%)
o
nanus
8/27
(29.6%)
c
scan 1 oni
3/25
(12.0%)
c
bauer i
2/24
(8.3%)
c
crepuscularis
3/3 (100%)
**
Degenerating oocysts present
Invasion of salivary glands by sporozoites

62
Table 7. Mean capture times for specimens of Cu 1 ico i des
taken in Bennett traps at Paynes Prairie and
Fisheating Creek. Values are in minutes,
relative to nautical sunset. Numbers in
parentheses are standard deviations.
Cu 1 i coi des Species - Paynes Prairie
Quarter*
hinma ni
eden i
a rborico1 a
1
18.7a
26.3a
(26.9)
(22.0)
2
-35.3a
15. lb
8.1c
(28.1)
(21.4)
(31.2)
3
-28.0a
12.0b
17.3b
(30.8)
(27.9)
(20.0)
4
-24.6a
14.8b
40.4c
(24.1)
(22.5)
(20.4)
Cu 1 icoi des
Species -
Fisheating Creek
Quarter
hinma ni
eden i
a rboricol a
knowltoni
1
-33.3a
5.1b
21.7bc*
51.5c*
(22.8)
(29.1)
(9.1)
(4.9)
2
-58.6a
0.0b
21.3c
26.9c
(32.9)
(35.3)
(31.1)
(18.2)
3
-49.3a
6.4b
24.8b*
26.5b
(27.6)
(32.9)
(41.4)
(16.8)
4
-23.9a
14.5b
S5.0c*
31.7bc
(27.8)
(29.4)
(19.1)
(20.4)
a Values with the same letter are not significantly
different, P< 0.05
* N 10
+ Quarter 1 = January-March; Quarter 2 = April-June;
Quarter 3 = Ju1y-Septembet; Quarter 4 = October-December

63
Table 8. Yearly prevalence of Haemoproteus me 1eaeridis
in specimens of Cu 1 ic o id e s e d e ni at Paynes
Prairie.
Month
Year
Poo 1 s
# Flies
Isolations
Apr i 1
1983
1
35
0
June
1983
1
83
0
July
1983
1
19
0
August
1983
1
10
0
November
1983
1
20
1
March
1984
1
12
0
April
1984
1
12
0
May
1984
2
152
1
Total :
9
343
2
Minimum YearIy Prevalence:
0.58%

64
Table 9. Yearly prevalence of Haemoproteus me 1eagridis
in specimens of Cu 1 icoi des edeni at H is heati ng
Creek.
Month
Year
Poo 1 s
# Flies
Isolations
April
1983
2
98
2
May
1983
2
23
0
June
1983
1
17
0
July
1983
4
99
3
Augus t
1983
2
42
0
September
1983
2
56
2
November
1983
7
101
I
December
1983
2
81
0
January
1984
1
9
0
February
1984
4
42
1
March
1984
9
117
5
April
1934
7
75
1
May
1984
6
56
2
Total:
49
816
17
Minimum Yearly Prevalence:
2.08%

65
Table 10. Attempted isolations of Haemoproteus me 1 eagridis
from pools of Cu I icoi dFs h i nman FT Cu i i co i de s
a r bo rico1 a and Cu 1 icoi des knowlton i it Paynes
Prairie ÍPAP) and Fisheating Creek (FEC).
Cu 1 icoi des hinmani
Location
Month
Year
Pool s
it Flies
Isolations
PAP
July
1983
1
3
0
PAP
August
1983
1
4
0
PAP
September
1983
1
8
0
FEC
May
1983
1
10
0
FEC
Augus t
1983
1
2
0
FEC
March
1984
1
7
0
FEC
April
1984
I
5
0
Culicoides arboricola
Location
Month
Year
Pool s
it Flies
Isolations
PAP
April
1983
1
10
0
PAP
May
1983
1
13
0
PAP
ivia r c h
1984
1
5
0
FEC
April
1984
1
8
0
Culicoides knowltoni
Location
Month
Year
Pool s
it Flies
Isolations
FEC
June
1983
1
3
0
FEC
July
1983
1
4
0
FEC
September
1983
1
4
0
FEC
May
1984
2
53
0

Figure 1
Figure 2
Figure 3.
Figure
Figure 5
Figure
Ookinete of Haemoproteus me 1eagr i d i s from the
midgut of a specimen ol Cu 1icoides edeni, 24
hours after the fly engorged on ah infected
turkey. A mass of pigment (arrow) is located
near the posterior end of the organism. Bar = 10
um. Note: Figures 1-6 were taken with Normarski
contrast interference microscopy.
Developing oocysts (arrows) of Haemoproteus
me 1 eagr i d i s on the midgut of a spec ¡men oT
Cu 1 i coi des eden i , 4 days after the fly had taken
a blood meal from an infected turkey. Bar = 50
um.
A 6-day-old, degenerating oocyst of Haemoproteus
me 1 e ag ridis from a specimen of Cu 1 icoide¥
edeni. The oocyst contains large retractile
granules (arrow). Bar = 10 um.
A mature, 6-day-old oocyst of Haemoproteus
me 1e a g r i d i s from a specimen of Cu 1 i coi des
e d e n i . The oocyst is packed with slender
sporozoites that are parallel to one another. A
small, retractile residual body (arrow) is
present. Bar = 10 um.
One of the 2 salivary glands from a specimen of
Cu 1 i coi de s e d e ni that had engorged on an
infected turkey 7 days earlier. The gland
consists of an elongate primary lobe and several
smaller, apical lobes. The shadows of
sporozoites are barely visible in secretory
cells composing the primary lobe (arrows). Bar
= 10 um.
A crushed salivary gland from a specimen of
Cu 1 icoi de s ed e ni that had engorged on an
infected turkey 7 days earlier. Numerous
elongate sporozoites (arrows) are located within
the secretory cells. Bar = 10 um.

67

Figure 7. Scatter plot of capture times for specimens of
C u 1 i c o i d e s hinma ni that were captured in Bennett
traps at P a y n e s P rairie and Fisheating Creek.
Capture time is plotted as minutes before or after
nautical sunset (reference line). Arrow indicates mean
capture time.


Figure 8. Scatter plot of capture times for specimens of
Cu 1 icoi des edeni that were capture in Bennett traps
at Paynes Prairie and Fisheating Creek. Capture time is
plotted as minutes before or after nautical sunset
(reference line). Arrow indicates mean capture time.

240
220
200
180
160
140
m 120
5
| 100
80
60
40
20
0
Culicoides edeni
X= 10.25
-140 -120 -100 -80 -60 -40 -20 0 * 20 40 60 80 100 120 140
CAPTURE TIME

Figure 9. Scatter plot of capture times for specimens of
Cu 1 iciodes arborico1 a that were captured in Bennett
traps at Paynes FFairie and Fisheating Creek.
Capture time is plotted as minutes before or after
nautical sunset (reference line). Arrow indicates
mean capture time.

10
9
8
7
6
5
4
3
2
I
0
Culicoides orboricolo
• •••
• • • •
X = 19.2
• • «s» •
•• • •
«8 •«*» • •
-140 -120 -100 -80 -60 ^40 -20 0 20 40 60 80 100 120 140
CAPTURE TIME

Figure 10. Scatter plot of capture times for specimens of
Cu I icoi des knowltoni that were captured in Bennett traps
at Paynes Prairie and Fisheating Creek. Capture time is
plotted as minutes before or after nautical sunset
(reference line). Arrow indicates mean capture time.

16
15
14
13
12
II
10
9
8
7
6
5
4
3
2
I
0
Culicoides knowltoni
• •
X = 28.3
• 0
#
+ •
0 0
0 0
00 00 m
üÉLJL
j i i i
•140 -120 -100 -80 -60 -40 - 20 0 20+ 40 60 80 100 120 140
CAPTURE TIME

Figure 11. Modified New Jersey suction trap catches of
specimens of Cu 1 icoi de s ed en i at Fisheating
Creek during tlie March, April and May, 1983
collecting trips. (#—• = canopy trap;0"'0 =
ground trap). Rising and setting suns indicate
the 2-hour sampling period that included
dawn or dusk.

LOG |o (N -i-1) LOG |o C N -t-1) LOG,q(N + I)
77
Culicoides edeni

Figure 12. Modified New Jersey suction trap catches of
specimens of C u1 ic o i d e s hinma ni a t
Fisheating Creek dur ing the Ma r cE~¡ April and
May, 1 983 collecting trips. (®—9 = canopy
trap; O—O = ground trap). Rising and setting
suns indicate the 2-hour sampling period
that included dawn or dusk.

LOGlo (N + I) LOGlo (Nt I) LOG10(Ntl)
79
17 19 21 23 01 03 05 07 09 II 13 15
HOUR

Figure 13. Transmission vs. abundance of 3 species of Cu 1 icoi des at
Site A at Paynes Prairie that were able to support
development of Haemoproteus me 1eagridis. (®--® =
Bennett trap catch; Q—.O ~= New Jersey light trap
catch). The kite diagram at the top of the figure
indicates the % of sentinel birds that became
infected with Haemoproteus me 1eagridis during 4-week
periods between May, 19 8 2 and June, 1 984 . Blank
areas in the diagram indicate periods where transmission
did not occur. Marks on the x-axis that follow each
month indicate the middle of that month.

LOG|0 (Nt I) LOG |o (Nil)

Figure 14. Transmission vs. abundance of 3 species of Cu I icoi des at
Site B at Paynes Prairie that were able to support
development of llaemoproteus me 1e a g ridisâ–  =
Bennett trap catch; O—O ~= New Jersey light trap
catch). The kite diagram at the top of the figure
indicates the % of sentinel birds that became
infected with Ha emo proteus me Ie a g rid i s during 4-week
periods between July, 1982 and June, 1984. Blank areas
in the diagram indicate periods where transmission
did not occur. Marks on the x-axis that follow each
month indicate the middle of that month.

LOG|Q (N+ I) LOG ¡o IN * II

Figure 15. Departures from normal for average monthly
temperatures during 1982, 1983 and 1984 at
Paynes Prairie. Normals are based on 30
year averages for each month between 1951
and 1980 (Climatological Data: Florida,
1982, 1983, 1984).

8S
MONTHS

Figure 16. Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Paynes Prairie. Normals are based on averages
of monthly totals between 1951 and 1980
(Climatological Data: Florida, 1982, 1983,
1984).

MONTHS
INCHES INCHES INCHES

igure 17. Average monthly temperatures at Paynes Prairie
( ® -ft ) and Fish eating Creek ( O O )
during 1982, 1983 and 1984 (Climatological
Data: Florida, 1982, 1983, 1984).

90
80
70
60
50
40
90
80
70
60
50
40
90
80
70
60
50
40
89
1982
i iii ii i I I I l 1
J FMAMJ JA S 0 N D
1983
i i i i i i I I i l I l
J FMAMJ J ASONO
1984
_i i i i i i i i 1 1 1 1
JFMAMJJ ASOND
MONTH

Figure 18. Transmission vs. abundance of 4 species of
Cu 1 i c oid e s at Fisheating Creek that were
able to ¡Up port development of Haemop r oteus
me I eag r i d i s. (#—© = Bennett trap catch; 0--0 =
New Jersey light trap catch). The kite diagram
at the top of the figure indicates the % of
sentinel birds that became infected with
Ha emoproteus me 1e ag rid i s during monthly
collecting FTips between March, 1983 and
September, 1984. Sentinels were not exposed
during the October, 1984 trip. None of the
sentinels exposed during the November, 1984
collecting trip became infected. Marks on the
x-axis that follow each month mark the
middle of that month.

91
~ 2
z
o
0
V'
DEC
JAN 1 fEB 1 MAR * APR 1 MAY
JUN 1 JUL 1 AUG ‘ SEP 1 OCT 1 NOV 1 OEC 1 JAN 1 F£B 1 MAR
AFR 1 MAY 1 JUN 1 JUL
1 AUG 1 SEP 1 OCT 1 NOV 1 DEC 1 JA
2
82
1983
11984
Culicoides hinmani
o 1
/
LOG
y'j
v /
\
V/y' a\
0
—o''
. . /
V '\
DEC
JAN 1 FEB 1 MAR1 APR1 MAY
JUN 1 JUL 1 AUG 1 SEP 1 OCT 1 NOV 1 OEC 1 JAN 1 FEB 1 MAR '
APR 1 MAY 1 JUN 1 JUL 1
AUG 1 SEP 1 OCT 1 NOV 1 OEC 1 JAN
821
1983
' 1984
'85
/ ’
/ \
Culicoides knowltoni
: \
! \
,6 \
p' ^ i
P'"’9,
/ \ / r \
/
q ! \f f
.
/' /
OEC
JAN 1 FEB 1 MAR
APR 1 MAY 1 JUN 1 JUL
AUG 1 SEP 1 OCT 1 NOV 1 OEC 1 JAN 1 FEB 1 MAR 1 APR 1 MAY 1 JUN
JUL 1 AUG ' SEP 1 OCT 1 NOV 1 DEC 1 JA
82
1983
' 1984
'85
Culicoides arboricola
_ 2
O

Figure 19. Departures from normal for average monthly
temperatures during 1982, 1983 and 1 984- at
Fisheating Creek. Normals are based on 30 year
averages each mo nth between 1951 and 1980
(Climatological Data: Florida, 1982, 1983,
1984).

93
MONTHS

Figure 20. Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Fisheating Creek. Normals are based on
averages of monthly totals between 1951 and
1980 (Climatological Data: Florida, 1982, 1983,
1984).

MONTHS
INCHES INCHES INCHES

96
Pathogen icity
Experiment 1 - Pathology
Gross observations - spontaneous deaths. As early as
7 days post-infeetion (DPI) a number of subtle, behavioral
changes became evident among poults within the high dose
group. The birds stood with slightly drooped wings,
ruffled feathers and partially or completely closed eyes
and were less active than birds belonging to either the low
dose or control groups. Between 7 and 14 DPI, birds in the
high dose group developed a mild diarrhea which produced a
"pasty" vent. Their physical condition continued to
deteriorate and by 15 DPI, most birds exhibited lameness in
1 or both legs and severe depression. Most were emaciated,
dehydrated and anorexic.
On days 19, 20 and 22 DPI, 4 birds (33%) in the
high dose group died. The deaths occurred from 2 to 5 days
following the appearance of young gametocytes in the
peripheral circulation. At necropsy, the skeletal muscles
contained numerous fusiform, white cysts, up to 1.0 nm
in length and 0.5 mm in diameter. They were scattered
diffusely throughout the skeletal muscles, but were most
corrmon in the pectoral muscles. All were oriented parallel
to the muscle fibers. From 30 to 50% of the cysts were
discolored red by hemorrhage (Figure 21).

97
The 4 birds had a number of secondary lesions
unrelated to the muscle cysts. One bird had a large white
nodule, approximately 5 mm in diameter, that occupied
the posterior portion of 1 lung. Another bird had
thickened air sacs with scattered white plaques that
were about 5 urn in diameter. In all 4 birds, portions of
the gut were swollen and flaccid and the mucosa had
multifocal areas of discoloration. Reddish or black, tarry
mucoid material was occasionally present in the jejunum. A
bacterial culture from 1 bird was positive for
Salmone11 a enterid i t i s Group 3.
A randomly selected bird in the control group was
killed and necropsied at 22 DPI to determine whether the
cysts observed in the birds that died may have resulted
from contamination with Sarcocystis. Tissue cysts were not
evident in any skeletal muscles of this bird. All
organs and tissues were grossly normal.
At 27 DPI, cloacal swabs from 2 of 3 randomly selected
low dose birds, 2 of 3 randomly selected control birds and
3 of 5 randomly selected high dose birds were positive for
Salmone1 I a enter id i t i s Group B, serotype heidelberg.
Gross observations - surviving birds. At 8 weeks
post-infection, all surviving birds were necropsied. Nine
of the 12 low dose birds and 8 of the 8 high dose birds had
low numbers of fusiform, white cysts in the pectoral

98
muscles. They were 2-3 times larger and were more diffuse
than cysts observed in the 4 fatal infections. Cysts were
not detected in any of the 11 control birds.
All surviving birds had focal areas of reddened,
swollen mucosa and occasional petechial hemorrhages
throughout the length of the gut. None of the birds
exhibited clinical signs of salmonellosis, i.e. "pasty"
vent, bloody diarrhea or depression. The "pasty vents"
observed in high dose birds at 1 week PI resolved
spontaneously in surviving birds by 4 weeks PI. At the
termination of the experiment, cloacal swabs from 2 of
11 control birds, 1 of 12 low dose birds and 1 of 8 high
dose birds were positive for Salinone11 a enteriditis
Group B. Coccidian oocysts were not detected in pooled
fecal samples at either 4 or 8 weeks post-infection.
Incidental findings included the presence of numerous
white nodules, approximately 5 ¡ira in diameter, that were
scattered throughout the mesentery lining the abdominal
cavity of 1 control bird and 2 high dose birds.
Average wet weights of hearts and livers removed from
the birds at necropsy were not significantly different for
any group (Table 11). The difference in average spleen
weights among the 3 groups was highly significant (p<
.0002).

Table 11. Average organ* weights at necropsy
Group
N
Heart
(p = 0.112)
Liver
(p = 0.0641)
Spleen
(p=0.0002)
Control
1 1
0.443
1.70
0.093
Low
12
0.457
1.74
0.130
High
8
0.489
1.93
0. 163
Weights expressed as % of total body weight at necropsy

100
Microscopic observations - spontaneous deaths. The
most significant lesions found in tissues from the 4
spontaneous deaths were in the skeletal muscles. All 4
birds had numerous intact and degenerating, fusiform
me ga 1 oschizonts between the muscle fibers. The
schizonts ranged from 43 to 155 urn in diameter with a mean
diameter of 95.0 urn (n = 35, SD = 34.67). Depending on the
plane of section, they ranged up to 600 um in length. Most
mega 1 oschizonts were oriented parallel to the muscle
fibers, although a few were perpendicular. Many were
outlined by a thick, hyaline wall which occasionally
indented slightly to form node-like constrictions (Figures
22, 23).
Mature mega 1 oschizonts were packed with small,
spherical merozoites less than 1 um in diameter. Each
contained a small mass of chromatin. Some immature
mega 1 oschizonts contained irregularly shaped cytomeres
of different sizes that were located next to the cyst
wall. Chromatin masses were arranged around the periphery
of each cytomere. Other mega 1oschizonts were packed
with round, granular, dark blue cytomeres. As
mega 1oschizonts matured, the number of cytomeres increased
in inverse proportion to their diameter. Merozoites
developed as buds from the outer surface of the cytomeres.
A severe hemorrhagic myositis was associated with the
mega 1 osch i zonts . Ruptured schizonts were surrounded by

101
a mixed inflammatory infiltrate composed of macrophages,
heterophils, giant cells and red blood cells (Figures
23, 28). They were frequently invaded by macrophages
and heterophils. Macrophages were often adjacent to the
outer wall of ruptured mega 1 os ch i zon t s and occasionally
were seen phagocytizing merozoites that had been liberated
(Figures 24, 25). Giant cells commonly were found adjacent
to intact and degenerating cysts (Figures 23, 25). Muscle
fibers adjacent to the cysts were swollen, rounded and
hyaline and often contained small, gray to dark blue
granules which were oriented occasionally into parallel
lines (Figures 26, 27). Dark blue calcium deposits,
visible with hematoxylin and eosin as well as with von
Kossa's calcium stain, occupied much of the cytoplasm of
muscle fibers adjacent to more degenerate cysts (Figures
26, 27).
Capillaries and venules adjacent to or near
degenerating mega 1oschizonts were often occluded partially
or completely by thrombi composed of pink staining,
fibrinous material (Figure 29). Giant cells often
surrounded thrombi or were adjacent to occluded vessels.
Liver and spleen sections from the 2 birds that
died spontaneously at 22 DPI, 5 days after gametocytes
first appeared in circulating red cells, had numerous
golden yellow pigment deposits in the cytoplasm of
macrophages (Figure 30). The 2 birds that died on 19

102
and 20 DPI lacked similar deposits. Spleen sections
from all 4 birds had extensive areas of follicular
hypoplasia caused by a large reduction in the size of
the periarterial lymphatic sheaths. Numerous mature and
ruptured schizonts were present in reticular cells in
the spleen of 1 bird. These schizonts were smaller in
diameter and lacked the thick, hyaline wall that surrounded
megaloschizonts from muscle tissue (Figures 31, 32). Most
averaged from 10 to 15 urn in diameter and contained small,
spherical zoites, less than 1 urn in diameter. A few
contained elongate zoites.
Heart sections from 1 bird had several thrombi with
associated giant cells and a single, large mega 1oschizont.
Incidental findings in lung tissue from 3 of the 4
birds included the presence of large granulomas composed of
giant cells, mononuclear cells and heterophils that
surrounded large, amorphous eosinophilic central cores.
Fungal hyphae resembling As pe r gi 11u s sp. were evident
in sections from 1 bird. Epithelial cells lining alveolar
capillaries were hypertrophic and associated capillaries
were congested. A f i brino-hemorrhag i c exudate containing
macrophages and heterophils flooded alveolar spaces. Some
large blood vessels were occluded by thrombi. Sections of
air sacs of 1 bird and mesentery of another had similar
granulomas that surrounded eosinophilic masses
containing fungal hyphae.

103
Focal aceas of enteritis characterized by the presence
of heterophils in the lamina propria and submucosa were
present in sections of intestine and cecum. Coccidian
parasites were not detected. Sections of kidney, brain,
bone marrow, proventri cu 1 us and gizzard were unremarkable.
Microscopic observations - surviving birds. At 8
weeks post-infection, nodular infiltrates of mononuclear
cells, macrophages, heterophils and giant cells were
evident in sections of skeletal muscle from low dose and
high dose birds (Figure 33). The hyaline remnants of
the outer wall of degenerating mega 1oschizonts and necrotic
and calcified muscle fibers were at the center of some
of the nodules. A scattered lymphocytic, heterophilic
infiltrate was frequently perivascular and often present
between muscle fibers. Pectoral muscle from 1 low dose
bird contained a degenerating cyst with merozoites (Figure
34). Thrombi surrounded by macrophages and heterophils
were present in some sections (Figure 34). Remnants of
degenerating muscle fibers, infiltrated with macrophages
and heterophils, were scattered randomly throughout the
sections of muscle (Figure 35).
Sections of liver, lung and spleen from both low
and high dose birds contained moderate to extensive, random
deposits of pigment. No pigment was found in control
birds. Deposits in the liver and spleen were massive

104
and brownish-black. Those in the lung were smaller and
golden-brown. All were contained in macrophages.
Follicular hyperplasia was corrmon in the spleens of all
infected birds. The degree of hyperplasia as well as
the number of pigment deposits varied directly with the
size of the infective dose.
Sections of gut from the control, low dose and high
dose birds had a few multifocal areas of infiltrate
composed of heterophils. Coccidian parasites were not
observed. One high dose bird had a granulomatous
peritonitis. The granulomas were composed of macrophages,
heterophils and giant cells that surrounded amorphous,
eosinophilic masses.
Sections of brain, bone marrow, kidney, heart,
proventri cu 1 us and gizzard were unremarkable.
Parasitemia. Young gametocytes appeared in the
peripheral circulation of all birds in the high and low
dose groups at 17 DPI. The control birds remained
uninfected throughout the study. The parasitemia in
both infected groups quickly reached a peak by 21 DPI
and then rapidly fell within 7 days to values less than 10%
of those at the crisis (Figure 36). A second, smaller peak
occurred at approximately 5 weeks pos t-infection . Both
groups remained patent through the course of the study,
although parasitemias were often less than 1.0 %.

105
At the crisis, the peak parasitemia in the high
dose birds reached an average high of 5,760 gametocytes per
10,000 red cells. Two birds had peak parasitemias that
exceeded 7,000 gametocytes per 10,000 red cells. In
many birds, more than 50% of the red cells contained
developing gametocytes. Multiple infections of red
cells were common, with some cells containing as many as 6
gametocytes. At the crisis, low dose birds had an average
peak parasitemia of 2,109 gametocytes per 10,000 red cells,
less than half that of the high dose group.
Weight. Statistical analysis of the weight data
revealed that all 4 variables in the model statement,
i.e. treatment, subject(treatment), week,
treatment*week, were highly significant (p< .0001).
When comparisons were made by week, all 3 groups were
significantly different 1 week post-infection (PI) and
at the crisis at 3 weeks PI. Weights of high dose birds
were significantly lower than control and low dose birds
during all other weeks. Other differences between the
control and low dose groups were not significant (Figure
37).
When comparisons were made within groups, control and
low dose birds had significant increases in weight at each
week PI. By contrast, average weights for the high dose
group increased during each week, but not significantly at

106
0, 1 and 2 weeks PI, at 1, 2 and 3 weeks PI and at 3 and 4
weeks PI (Figure 37).
Tarsometatarsal length. Statistical analysis of
tarsometatarsal length revealed that all 4 variables in the
model statement were highly significant (p = .0001). When
comparisons were made by week, all 3 groups were
significantly different at the crisis, at 3 weeks PI. The
high dose group had average tarsometatarsal lengths that
were significantly shorter than control and low dose birds
at all other weeks PI. Other differences between the
control and low dose groups were not significant (Figure
38).
When comparisons were made within group, all groups
showed a significant increase in tarsometatarsal length for
each week.
Hema tocrit. Statistical analysis of hematocrit
revealed that all variables in the model statement were
highly significant (p = .0001) except treatment*week (p
= .6886). Comparisons among the 3 experimental groups,
averaged over all weeks, were not significant (p = .1617).
Comparisons among weeks, averaged over all groups,
showed no significant differences among weeks 0, 6 and
8, among weeks 3, 6 and 7, among weeks 1, 4 and 5 and
between weeks 2 and 4 PI (Figure 39).

107
Plasma protein concentration. Statistical analysis of
plasma protein concentration revealed that all variables in
the model statement were significant (p < .0125).
When comparisons were made by week, all 3 groups were
significantly different at I week PI. Control birds had
the highest average plasma protein concentration and
high dose birds had the lowest. 3y 2 weeks PI, average
plasma protein concentrations were significantly greater
for high dose birds than either low dose or control birds.
The 3 experimental groups were not significantly different
at 3 and 4 weeks PI. At 5 and 6 weeks PI, high dose birds
had average plasma protein concentrations that were
significantly higher than low dose and control birds. The
3 groups were not significantly different at 7 and 8 weeks
PI (Figure 40).
When comparisons were made within groups, the low dose
birds had significantly lower average plasma protein
concentrations at weeks 0, 1 and 3 PI than they did at
weeks 2, 4, 5, 6, 7 and 8 PI.
Average plasma protein concentrations in the high dose
group were significantly lower at 1 and 3 weeks PI than
they were at 0, 4, 7 and 8 weeks PI. Average
concentrations at the latter 4 weeks were significantly
lower than those at 2 and 5 weeks PI. Average values at 0
and 3 weeks PI were also significantly lower than those at
4, 5, 6, 7 and 8 weeks PI.

108
When comparisons were made within the control group, a
considerable amount of overlap was detected.
Significant differences were not detected among 2, 3, 4, 6
and 7 weeks PI, among 1, 2, 4, 6, 7 and 8 weeks PI and
among 1, 2, 5, 6 and 8 weeks PI. Average values at week 0
were significantly lower than those at any other week PI.
Hemog1obin. Statistical analysis of hemoglobin
data revealed that all 4 variables in the mo d e 1
statement were highly significant (p = .0001).
When comparisons were made by week, high dose birds
had significantly lower average hemoglobin values at 4
weeks PI than either low dose or control birds. No
differences were significant at other weeks PI.
When comparisons were made within the low dose group,
average values at 6, 7 and 8 weeks PI were significantly
lower than the average at week 0. Average values at week 0
were significantly lower than those at 3 and 4 weeks
PI. Average values at 3 and 4 weeks PI were significantly
lower than those at 1, 2 and 5 weeks PI (Figure 41).
Comparisons within the high dose group had
considerable overlap. Average values at 4, 6, 7 and 8
weeks PI were significantly lower than those at 1, 3 and 5
weeks PI. Average values at 0, 3 and 7 weeks PI were
significantly lower than those at 1, 2 and 5 weeks PI.
Significant differences were not detected among 0, 3 and 7

109
weeks PI, among 0, 3 and 5 weeks PI and among I, 2 and 5
weeks PI.
Average hemoglobin values within the control group
were significantly lower at 6, 7 and 8 weeks PI than
they were at week 0. Average values at week 0 were
significantly lower than those at 3, 4 and 5 weeks PI.
Average values at 3, 4 and 5 weeks PI were significantly
lower than those at 1 and 2 weeks PI.
Experiment 2 - Exoerythrocyt i c Development
Three days. At 3 days post-infection, sections of
skeletal muscle from the inoculated poult had a few
focal areas of perivascular monocytic infiltrate. Some
muscle fibers appeared disrupted, with granular sarcoplasm
and scattered islands of pale, disorganized myoglobin
(Figure 53). A single, uninucleate parasite with blue-gray
granular cytoplasm was detected within a capillary (Figure
42). The organism was 3 urn in diameter and present in only
1 of a series of 4 um serial sections. Other tissues in
the infected and control bird were unremarkable.
Five days. By 5 days post-infect ion, infected poults
developed severe lameness in both legs. All had difficulty
standing and moving about their cage, while control
birds remained active. Skeletal muscle from an infected
bird had focal areas of necrosis that involved many

110
adjacent muscle fibers. The fibers were swollen, pale and
hyaline and often had disrupted sarcoplasm (Figure 54).
Individual swollen, rounded and hyaline muscle fibers were
scattered randomly throughout healthy tissue. A few focal
areas of perivascular monocytic infiltrate were present.
Schizonts measuring from 12 to 20 um in diameter were
present both within and between muscle fibers (Figures 43,
44). The smaller, more immature parasites contained
dark-staining, granular cytoplasm (Figure 43). The larger,
more mature forms were packed with dark-staining, elongate
zoites that were bent and twisted around each other (Figure
44). None of the parasites had an associated host
reaction.
Sections of heart from the infected bird had focal
aggregates of mononuclear cells scattered randomly
throughout the tissue. Hepatocellular atrophy and necrosis
were evident in sections of liver. The spleen was enlarged
and contained numerous erythropoietic cells in the vascular
sinuses. Other tissues from the infected and control bird
were unremarkable.
Eight days. By 8 days post-infection , infected birds
showed dramatic signs of improvement and were back on their
feet. Sections of skeletal muscle from 1 infected bird had
numerous focal areas of necrotic muscle fibers, infiltrated

by macrophages and giant cells. Regenerating muscle fibers
were corrmon (Figure 55).
Small schizonts with blue-gray cytoplasm and numerous,
irregularly shaped black nuclei were present in
capillaries. The schizonts were sausage shaped, with an
irregular outline and ranged from 5 to 8 um in diameter up
to 28 um in length (Figures 45, 46). Most schizonts
extended only as far as 2 to 3, 4 um serial sections. Some
schizonts were found adjacent to or within the muscle
lesions, although most showed no evidence of an associated
host response.
Hepatocellular atrophy and necrosis were evident in
sections of liver from the infected bird. Numerous large,
10 - 20 um, intracellular vacuoles were scattered
throughout the tissue. The spleen was enlarged, but
otherwise unremarkable. Other tissues from the infected
and control bird were unremarkable.
El even days. At 11 days post-infection, infected
birds continued to improve. Regenerating muscle fibers,
often surrounded by a monocytic infiltrate, were common in
sections of skeletal muscle from I infected bird. Necrotic
areas containing giant cells and macrophages were rare.
Perivascular nodular infiltrates composed of monocytes and
some heterophils were present. Other tissues from the
infected and control bird were unremarkable.

112
Schizonts were not detected in skeletal muscle or
other tissues in the infected bird.
Fourteen days. By 14 days post-infection, infected
birds were smaller than controls, but were otherwise
indistinguishable. Regenerated muscle fibers were
common in sections of skeletal muscle from 1 infected
bird. The regenerated areas occasionally contained
remnants of necrotic fibers that were surrounded by a
monocytic infiltrate. Perivascular, nodular infiltrates
composed of monocytes were present.
Schizonts ranged from 20 to 32 um in diameter and
developed both within and between muscle fibers (Figures
47, 48, 49). They were surrounded by a thick, hyaline wall
and were packed with dark blue, granular cytomeres that
ranged from 2 to 3 um in diameter. Schizonts were elongate
and extended as far as 90 um along the long axis of the
muscle fibers. Several schizonts were densely packed with
dark-staining granules (Figure 47). Others were surrounded
by macrophages (Figure 49).
Sections of heart from the infected bird had focal
areas of mononuclear infiltrates. Other tissues from
the infected and the control bird were unremarkable.
Seventeen days . At 17 days post-infection,
pectoral muscle from the last infected bird contained a few

113
areas of diffuse, white streaks, several rnn in length,
embedded deeply in the tissue. These were less distinctive
than the white fascia that separated muscle bundles. Fully
mature megaloschizonts containing numerous, densely packed,
spherical zoites and several large, central vacuoles
were present in sections of the tissue (Figure 50, 51).
Immature forms resembling 14-day-old schizonts were also
present. The mega losch i zon t s were surrounded by a thick
hyaline wall and ranged from 30 to 113 urn in diameter.
They extended as far as 465 urn along the long axis of
muscle fibers. Fibers adjacent to the mega 1oschizonts were
swollen, pale and hyaline (Figure 50). Several layers
of connective tissue, infiltrated with macrophages,
surrounded some schizonts. A large degenerating
meg a 1 oschizont , filled with amorphous, gray material
containing irregularly shaped red masses, was present in 1
section. The mega 1oschizont was surrounded by giant cells
and an outer layer of connective tissue (Figure 56).
Ruptured, partially empty mega 1oschizonts that contained
small numbers of scattered zoites were also present (Figure
52). Each zoite contained a small mass of chromatin and a
large vacuole. Giemsa-sta i ned erythrocytes from the
same bird contained young gametocytes that were
morphologically indistinguishable from the exoerythrocytic
zoites.

114
Sections of heart from the infected bird had focal
areas of monocytic infiltrate. A single mega 1 oschizont
without an associated host response was present. Other
tissues from the infected and control birds were
un remarkab1e.
Exoerythrocytic Development in Natural Infections
In November, 1970, a male Wild Turkey was captured
near Lake Apopka, Florida. The bird died soon after
capture, was frozen and then necropsied several weeks later
by K.P.C. Na i r and D.J. Forrester (pets. comm.). At
necropsy, scattered whitish cysts, the size of millet
seeds, were noticed on the pectoral muscles. Histological
examination of the tissue revealed occasional spherical
mega 1 oschizonts from 200 - 400 um in diameter. The
mega 1oschizonts were surrounded by a thick hyaline wall
(Figures 57, 58). Some contained disorganized masses of
dark staining material (Figure 57). Others held
numerous spherical merozoites (Figure 58). Muscle
fibers surrounding some of the mega 1oschizonts were
pale, swollen and had hyaline cytoplasm. Some fibers
contained small, dark-staining granules (Figure 58).
Other infections with the organism were not
detected in many subsequent necropsies of Wild Turkeys from
northern and southern Florida (Forrester, pers. comm.).

Figure 21. Formalized portion of pectoral muscle from a
high dose bird that died spontaneously at 19
days post-infection. Mega 1oschizonts appear as
numerous white streaks (arrows) scattered
throughout the tissue. The dark flecks and
discolored areas are hemorrhagic cysts. Bar =
1 cm.

116

Figure 22. An intact mega 1osch i zont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection .
Node-like constrictions (arrows) occur along
its length. Hematoxylin and eosin. Bar =
1 OOum.
Figure 23. An intact mega 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection .
The mega 1 oschizont is surrounded by a thick,
hyaline wall (arrows). The interior is packed
with minute merozoites. Cytomeres are not
evident. The megaloschizont is surrounded by a
mixed inf1anrnatory infiltrate composed of giant
cells (double arrow), macrophages, heterophils
and red blood cells. Hematoxylin and eosin.
Bar = 50 um.

118

Figure 24.
Figure 25.
A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection .
Hyaline, necrotic muscle fibers (large
arrow) and a mononuclear infiltrate (double
arrow) surround the schizont. Hematoxylin and
eos in. Bar = 50 um.
A degenerating mega 1oschizont from pectoral
muscle of a high dose bird that
spontaneously at 19 days post-infection .
The cyst contains numerous mononuclear cells
(arrow) and a fibrinous exudate. Giant
cells (double arrow) are adjacent to 1 side of
the cyst. Hematoxylin and eosin. Bar = 50 um.

120

Figure 26. An intact meg a 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection .
The megaloschizont is surrounded by
dark-staining areas of calcification. Parallel
lines of dark-staining granules are present in
1 muscle fiber adjacent to the schizont
(arrow). Hematoxylin and eosin. Bar = 50 um.
Figure 27. A serial section of the megaloschizont
illustrated in Figure 26. The section is
stained with von Kossa's stain for calcium.
There is a close correspondence between the
dark-sta ining deposits in Figure 26 and the
dark-sta ining calcium deposits in Figure
27. Bar = 50 um.

122

Figure 28. A degenerating mega 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-in£ection .
The hyaline cyst wall (arrow) and a mixed
inflammatory infiltrate composed primarily
of red blood cells is all that remains of
the schizont. Hematoxylin and eosin. Bar = SO
um.
Figure 29. A venule blocked by a thrombus in pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-infection.
The thrombus is surrounded by a giant cell
(arrow). Hematoxylin and eosin. Bar = 50 um.

124

Figure 30. A spleen section from a high dose bird that
died spontaneously at 22 days pos t-i n £ ec t i on .
Numerous masses of dark pigment are
scattered throughout the tissue.
Hematoxylin and eos in. Bar = 20 urn.
Figure 31. A schizont in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection. The host cell has a
pycnotic nucleus (arrow). The schizont is much
smaller than megaloschizonts from skeletal
muscle in the same bird and lacks a thick,
hyaline wall. Hematoxylin and eosin. Bar = 10
urn.
Figure 32. Schizonts in the spleen of a high dose bird
that died spontaneously at 22 days
post-infection. One schizont contains elongate
zoites (large arrow). Free merozoites
(small arrow) are scattered throughout the
tissue. Hematoxylin and eosin. Bar = 10 urn.

931

Figure 33. Nodular infiltrate from pectoral muscle of a
high dose bird that was killed at 8 weeks
post-infection. The infiltrate is composed
primarily of mononuclear cells and surrounds
some dark-sta ining areas of calcification
(arrow). Hematoxylin and eosin. Bar = 50 um.
Figure 34. A thrombus, composed of fibrinous material,
adjacent to the remnants of a degenerating
mega 1oschizont (arrow). The section is from
pectoral muscle of a high dose bird that was
killed at 8 weeks post-infection . Hematoxylin
and eosin. Bar = 20 um.
Figure 35. A mass of degenerating muscle fibers,
infiltrated with macrophages and heterophils.
Section is from pectoral muscle of a high dose
bird that was killed at 8 weeks
post-infection. Hematoxylin and eosin. Bar
= 20 um.

128

Figure 36. Average parasitemia for high dose birds ( O O ),
low dose birds ( ér~-*ék ) and control birds ( —® ).
The crisis occurred on the day when the peak parasitemia
was reached.

LOG|q (Parasitemia +1)
O — ro w ^
oe i
Crisis

Figure 37. Average weights of high dose birds ( O O
dose birds ( A--*-A ) and control birds ( #-
Statistically significant differences among
for each week of the study are indicated by th
points (p 0.05).
) , low
• ) .
groups
e boxed

4000
3500
3300
2500
2000
1500
1000
500
0
WEEKS
132

Figure 38. Average tarsometatarsal lengths of high dose birds
(Q"""0 ), low dose birds ( A A ) and control birds
(@ @ ). Statistically significant differences among
groups for each week of the study are indicated by
the boxed points (p 0.05).

134

Figure 39. Average hematocrits for high dose birds (
low dose birds ( A A ) and control birds
There were no statistically significant di
among groups when comparisons were made by
0.05).
O O ),
(© © ).
fferences
week (p

50
45
40
35
30
25
20
Crisis
I
-I 1 I J I I I I ■ - —I
01 2345678
WEEKS

Figure 40. Average plasma protein concentrations for high dose
birds (O O ), low dose birds ( A A ) and control
birds (# ® ). Statistically significant differences
among groups for each week of the study are indicated by
the boxed points (p 0.05).

WEEKS

Figure 41. Average hemoglobin concentrations for high dose birds
(O O ), low dose birds ( A A ) and control birds
( @ © ). Statistically significant differences among
groups for each week of the study are indicated by
the boxed points (p 0.05).

-101 2345678
WEEKS
Otl

Figure 42. Three-day-old schizont (arrow) from pectoral
muscle of an experimentally infected
turkey. The schizont is within a
capillary. Hematoxylin and eosin. Bar = 10
um.
Figure 43. Five-day-old schizont from pectoral muscle
of an experimentally infected turkey. The
schizont is located between muscle fibers
and is packed with dark granules. Hematoxylin
and eosin. Bar = 10 um.
Figure 44. Five-day-old schizont from pectoral muscle
of an experimentally infected turkey. The
schizont is located within a muscle bundle and
is packed with elongate zoites. Hematoxylin
and eosin. Bar = 10 um.
Figure 45. Eight-day-o 1d mega 1 oschizont (arrow) from
pectoral muscle of an experimentally
infected turkey. The megaloschizont is located
within a capillary and contains several
dark-sta ining nuclei. Hematoxylin and eosin.
Bar = 10 um.
Figure 46. Eight-day-o1d mega 1 oschizont from pectoral
muscle of an experimentally infected turkey. A
giant cell (double arrow) is adjacent to the
mega 1 oscli i zon t (arrow). Hematoxylin and eosin.
Bar = 10 um.
Figure 47. Fourteen-da y-o 1d me ga 1 oschizont from
pectoral muscle of an experimentally
infected turkey. The mega 1oschizont is located
within a muscle fiber and is surrounded by a
thick, hyaline wall (arrow). Hematoxylin
and eosin. Bar = 10 um.

142

Figure 48. Fourteen-day-old mega 1 oschi ton t from
pectoral muscle o£ an experimentally
infected turkey. The mega 1 oscli i zon t
contains dark, spherical cytomeres and is
surrounded by a thick, hyaline wall
(arrow). Hematoxylin and eosin. Bar = 10 urn.
Figure 49. Fourteen-day-old megaloschizont from
pectoral muscle of an experimentally
infected turkey. The megaloschizont
contains spherical, mu 1 t i nuc1eated cytomeres
and is surrounded by a hyaline wall (double
arrow) and a mononuclear infiltrate. A swollen
muscle fiber with disrupted, hyaline cytoplasm
is near the megaloschizont (arrow).
Hematoxylin and eosin. Bar = 10 um.
Figure 50. Seventeen-day-old megaloschizont from pectoral
muscle of an experimentally infected
turkey. The megaloschizont is surrounded by
several layers of connective tissue (CT) and
contains vacuolated areas (V). Swollen,
pale and hyaline muscle fibers (arrows) are
adjacent to the megaloschizont. Hematoxylin
and eosin. Bar = 50 um.


Figure SI. Seventeen-day-old mega 1oschizont from pectoral
muscle of an experimentally infected
turkey. The megaloschizont contains
numerous elongate and branching cytomeres with
budding merozoites. Hematoxylin and eosin.
Bar = 10 um.
Figure 52. Seventeen-day-old megaloschizont from pectoral
muscle of an experimentally infected
turkey. The thick, hyaline outer wall of
the megaloschizont (large arrow) has ruptured,
liberating the merozoites. The merozoites are
spherical and contain a large vacuole (small
arrow). Hematoxylin and eosin. Bar = 10 um.

146

Figure 53. Disrupted muscle fiber (arrow) from pectoral
muscle of a turkey with a 3-day-old
experimental infection of Ha emo proteus
me 1eagridis. Hematoxyl in and eosin. Bar =
10 um.
Figure 54. Swollen, hyaline and disrupted pectoral muscle
fibers from a turkey with a 5-day-old
experimental infection of Ha emo proteus
me 1 ea^r i d i s . Normal muscle fibers (M)sur round
the lesion. Hematoxylin and eosin. Bar =
10 um.
Figure 55. Regenerating muscle fibers (arrows) from
pectoral muscle of a turkey with an
8-day-old experimental infection of
Ha emo proteus me 1eag rid i s . Hematoxylin and
eosin.3ar = 10 um.
Figure 56. Deteriorating 17-day-old megalosch i zont (S)
from pectoral muscle of an experimentally
infected turkey. The mega 1 oschizont (S) is
surrounded by giant cells (G) and an outer
layer of connective tissue (arrow).
Hematoxylin and eosin. Bar = 100 um.


Figure 57. Mega 1 oschizonts from pectoral muscle of a
naturally infected Wild Turkey collected
near Lake Apopka, Orange County, Florida in
November, 1970. Mega loscli i zont s have a thick,
hyaline outer wall (arrows). Hematoxylin
and eosin. Bar = 50 um.
Figure S8. Mega 1oschizont from pectoral muscle of a
naturally infected Wild Turkey collected
near Lake Apopka, Orange County, Florida in
November, 1970. Mega 1 oschizont has a thick,
hyaline wall (arrow) and is surrounded by pale,
swollen and hyaline muscle fibers (M).
Hematoxylin and eosin. Bar = 50 um.

150
V
58

151
Host Spec ificity
Parasitemia
Haemoproteus me 1eagr i d i s was successfully transmitted
to 1 of 2 Ring-necked Pheasants, 1 of 2 Chuckars and 2
turkeys that acted as positive controls in the first series
of experimental infections. One of the 2 inoculated
Guineafowl was found dead 6 days post-infection (DPI). One
day prior to death, the bird appeared healthy. At
necropsy, all organs appeared normal. Histological
sections of liver, lung, heart, kidney, gizzard, pancreas,
duodenum, cecum and brain were unremarkable. Skeletal
muscle and spleen were not fixed. The surviving inoculated
Guineafowl and all negative controls failed to develop
patent infections throughout the duration of the
exper ¡merit.
All infected birds became patent at 17 DPI. The 2
infected turkeys reached an average peak parasitemia of 813
parasites per 10,000 red blood cells at 22 DPI. The
parasitemia rapidly fell by 28 DPI to 33 parasites per
10,000 red blood cells and remained at levels less than 15
parasites per 10,000 red blood cells for the duration of
the experiment (Figure 59). By contrast, the positive
Chuckar reached a lower peak parasitemia of 162
parasites per 10,000 red blood cells at 17 DPI. A
rapid drop to 23 parasites per 10,000 red blood cells

152
occurred by 20 DPI. Gametocytes were cleared from the
circulation by 36 DPI. The Ring-necked Pheasant that
developed a patent infection proved to be the least
susceptible to me 1eag r i d i s . The parasitemia reached
a peak of only 22 parasites per 10,000 red blood cells
at 20 DPI and was cleared from the peripheral
circulation by 26 DPI (Figure 59).
The 2 turkeys inoculated in the second series of
experimental infections were the only birds that developed
patent infections of me 1 e a g r i d i s â–  The Northern
Bobwhites, chickens and all negative control birds remained
negative throughout the course of the experiment.
Morphometric Analysis
Macrogametocytes. Macrogametocytes from each of
the infected host species were morphologically similar
(Figures 60, 62, 64). All completely encircled the host
cell nucleus and were consistent with descriptions of
neotypes of me 1eagridis (Greiner and Forrester, 1980).
The average adjusted cell area of macrogametocytes from the
turkey was larger than that of macrogametocytes from either
the Chuckar or the Ring-necked Pheasant (Table 12).
Average values of other adjusted variables were within 1
standard deviation of each other (Table 12).

153
A discriminant analysis was performed on adjusted
variables from a calibration data set derived from 15
macrogametocytes from the Chuckar, 15 from the turkey and 7
from the Ring-necked Pheasant. It derived a function that
correctly classified 80% of the discriminant scores from
the Chuckar and 80% of those from the turkey (Table
13). It correctly classified only 1 (14.3%) of the 7
scores from the Ring-necked Pheasant. Five of the pheasant
scores (71.4%) were incorrectly identified as turkey and 1
(14.3%) failed to meet criteria for classification in
any of the 3 categories (Table 13).
A small data set composed of adjusted measurements of
4 macrogametocytes from the turkey, 4 from the Chuckar and
3 from the Ring-necked Pheasant was analyzed to test the
validity of the derived function. The derived function
correctly classified 100% of the turkey scores, but only
25% of the Chuckar scores and 33% of the pheasant scores -
values close to what would be expected by chance alone
(Table 14). Three (75%) of the 4 Chuckar scores were
incorrectly classified as turkey (Table 14).
Microgaroetocytesâ–  Microgametocytes from each of the 3
host species were morphologically similar and encircled the
host cell nucleus (Figures 61, 63, 65). Average
adjusted values of cell area and cell length were smallest
for m i crogametocytes from the Ring-necked Pheasant

154
(Table 15). However, average cell width was greater in
mi erógametocyts from the pheasant and probably related
to the large, lateral displacement of the host cell nucleus
(Figure 65). The average adjusted number of pigment
granules in mi crogametocytes from the Chuckar was
considerably higher than values from the pheasant and
turkey.
A discriminant analysis was performed on adjusted
variables from a calibration data set derived from 15
m i crogametocytes from each host. It correctly
identified 73.3% of the discriminant scores from the
Chuckar, 60.0% of the scores from the pheasant and 66.7% of
the scores from the turkey (Table 16).
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 microgametocytes
from each of the 3 host species. It correctly identified 3
(75%) Chuckar scores, 3 (75%) turkey scores, but no (0%)
pheasant scores. Two pheasant scores were identified as
turkey, 1 as Chuckar and 1 failed to meet the criteria for
inclusion into any of the 3 categories (Table 17).
Infected host cells - macrogametocytes. Host cells
infected with macrogametocytes underwent a number of common
changes in each host species (Figures 60, 62, 64). All had
greater cell lengths and cell areas than corresponding
uninfected cells from the same host species. Infected host

1 55
cells from the turkey underwent the greatest enlargement.
Cells from all 3 hosts also had atrophied nuclei with
smaller lengths, widths and areas than nuclei from
unparasitized red blood cells (Table 18). Infected host
cells from the Ring-necked Pheasant underwent the greatest
nuclear atrophy.
A discriminant analysis was performed on adjusted
variables from a calibration data set composed of
measurements of 15 Chuckar cells infected with
ma crogametocytes , 15 infected turkey cells and 7
infected pheasant cells. It correctly identified 73.3% of
the Chuckar scores, 66.7% of the turkey scores and none of
the pheasant scores. Three (42.9%) of the pheasant scores
were identified as Chuckar and 3 (42.9%) were identified as
turkey - percentages approximately equal to what would
be expected by chance alone (Table 19). One pheasant score
failed to meet the criteria for inclusion in any of the
3 categories.
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 host cells from
the Chuckar and turkey and 3 host cells from the pheasant.
Three (75%) of the Chuckar scores were classified as turkey
and 2 (66.7%) of the pheasant scores were classified as
Chuckar. Two (50%) turkey scores were correctly
identified, but the classification was only slightly
less than would be expected by chance alone (Table 20).

156
Infected host cells - mi crogametocytes . Host cells
infected with mi erógame toeytes underwent several common
changes in each host species. All had an increase in
average cell length, width and area and a corresponding
decrease in nucleus length, width and area (Figures 61, 63,
65). A lateral displacement of the host cell nucleus
occurred in parasitized red cells from each host. The
d i s pI ac erne nt was greatest in host cells fr om the
Ring-necked Pheasant and least in host cells from the
Chuckar (Table 21). Differences in infected host cell
morphology were most evident in cells from the
pheasant. Infected red blood cells were rounder than those
from the other host species and the red blood cell nucleus
was often displaced to the outer margin of the host cell.
A discriminant analysis was performed on a calibration
data set composed of adjusted measurements of 15 host
cells, infected with mi crogametocytes, from each host
species. It correctly classified 60% of the Chuckar
scores, 46.7% of the pheasant scores and 60% of the turkey
scores (Table 22).
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 host cells from
each of the 3 host species. Two (50%) of the Chuckar
scores were classified as Chuckar, 1 (25%) of the turkey
scores was classified as turkey and 4 (100%) of the

157
pheasant scores were correctly classified as pheasant
(Table 23).
Fine Structure
Mature Gainetocytes
Mi erógametocytes and macrogametocytes were bound by a
pellicle composed of 3 unit membranes. The innermost
membrane was thicker and more osmiophilic than the outer 2
(Figures 66, 67). Other common organelles included a
nucleus bounded by 2 unit membranes, mitochondria with
tubular cristae and food vacuoles that contained
osmiophilic masses of pigment (Figures 66, 67). Both
macrogametocytes and microgametocytes had granular
nucleoplasm with an electron density similar to the
cytoplasm (Figures 66, 67). Macrogametocytes had an
electron-dense nucleolus (Figure 67).
Mature mactogametocytes were packed with numerous
ribosomes that gave the cytoplasm a granular appearance
(Figure 67). By contrast, mature microgametocytes
contained fewer ribosomes and had a paler, more
amorphous cytoplasm (Figure 66). Rough endoplasmic
reticulum and a Golgi apparatus were not observed.
However, gametocytes had a network of smooth endoplasmic
reticulum that extended throughout the cytoplasm. In

158
Table 12. Average adjusted measurements of
macrogametocytes. Average measurements of each
variable are expressed as a percentage of the
average uninfected host cell area for that
spec ies.
Var ¡able
Chuchar
Pheasant
Turkey
Parasite Length
43.2
(4.29)+
39.9
(1.59)
43.7
(2.47)
Parasite Width
4.4
(1.11)
3.6
(0.82)
4.9
(0.64)
Paras ite Area
97.9
(10.8)
104.3
(7.71)
117.2
(9.49)
Nucleus Length
8.8
(1.37)
7.9
(1.08)
7.2
(1.29)
Nucleus Width
4.1
(0.83)
3.3
(0.68)
4. 1
(0.59)
Nucleus Area
13.9
(2.36)
1 1 .4
(2.22)
13.3
(2.79)
P i gtne n t *
46.1
(14.4)
53.0
(5.55)
46.9
(6.31)
N =
15
7
15
+ Standard deviation
* Adjusted average of number of pigment granules.

159
Table 13. Classification summary of a nearest neighbor
analysis of a set of calibration data
composed of adjusted measurements of
macrogametocytes . Discriminant scores were
classified with a discriminant function derived
from the calibration data set summarized in
Table 12.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
12
(80.0)
+ 0
(0.0)
2
(13.3)
1 (6.7)
15
(100)
Pheasant
0
(0.0)
1
(14.3)
5
(71.4)
1 (14.3)
7
(100)
Turkey
2
(13.3)
0
(0.0)
12
(80.0)
I (6.7)
15
(100)
Total
14
(37.8)
1
(2.7)
19
(51.4)
3 (8.1)
37
(100)
Priors*
0.
,4054
0.
,1892
0
.4054
+ Percent of total
* Prior probability of being assigned to that class

160
Table 14. Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of macrogametocytes. The
discriminant function derived from data in Table
12 was used to classify the discriminant scores.
Classified into Species
Spec i es
Chuckar
Pheasant
Turkey
Other
Tota 1
Chuckar
1 (25.0)
+ 0 (0.0)
3 (75.0)
0
(0.0)
4
(100)
Pheasant
1 (33.3)
I (33.3)
1 (33.3)
0
(0.0)
3
(100)
Turkey
0 (0.0)
0 (0.0)
4 (100)
0
(0.0)
4
(100)
Total
2 (13.2)
1 (9.1)
8 (72.7)
0
(0.0)
11
(100)
Priors*
0.4054
0.1892
0.4054
+ Percent of total
* Prior probability of being assigned to that class

161
Table 15. Average adjusted measurements of
mi erógametocytes. Average measurements of each
variable are expressed as a percentage of the
average uninfected host cell area.
Variable
Chuckar
Pheasant
Turkey
Paras ite Length
41.4 (I.96)+
32.2 (5.65)
37.5
(6.01)
Parasite Width
4.6 (0.67)
6.1 (1.21)
5.8
(0.67)
Parasite Area
103.7 (9.03)
86.1 (8.38)
106.3
(10.6)
Pigme n t *
39.5 (12.0)
25.3 (5.88)
24.3
(4.72)
N =
IS
15
15
+ Standard deviation
* Adjusted average of number of pigment granules.

162
Table 16. Cl assi £ication sumnary o£ a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of mi crogametocytes.
Discriminant scores were classified with a
discriminant function derived from the
calibration data set summarized in Table 15.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Ch u c ka r
11
(73.3)+
2
(13.3)
2
(13.3)
0 (0.0)
15
(100)
Pheasant
0
(0.0)
9
(60.0)
5
(33.3)
1 (6.7)
15
(100)
Turkey
3
(20.0)
2
(13.3)
10
(66.7)
0 (0.0)
15
(100)
Tota 1
14
(31.1)
13
(28.9)
17
(37.8)
1 (2.2)
45
(100)
Priors*
0.
3333
0.
3333
0
. 3333
+ Percent of total
* Prior Probability of being assigned to that class

163
Table 17. Classification sumnary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of mi erógametocytes.
The discriminant function derived from data
in table 15 was used to classify the
discriminant scores.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Ch u c k a r
3
(75.0)
+ 0
(0.0)
1
(25.0)
0 (0.0)
4
(100)
Pheasant
1
(25.0)
0
(0.0)
2
(50.0)
1 (25.0)
4
(100)
Turkey
1
(25.0)
0
(0.0)
3
(75.0)
0 (0.0)
4
(100)
Total
5
(41.7)
0
(0.0)
6
(50.0)
1 (8.3)
12
(100)
Pr iors*
0.
.3333
0.
3333
0.
.3333
+ Percent of total
* Prior probability of being assigned to that class

164
Table 18. Average adjusted measurements of host cells
infected with macrogametocytes. Each variable
was divided by the average value of the same
variable from uninfected cells of the same
spec i es.
Variable
Chuckar
Pheasant
Turkey
Ce 11 Length
1.12
(0.08)+
1.10
(0.05)
1.09
(0.08)
Cell Width
1.02
(0.08)
1 .00
(0.07)
1.20
(0.08)
Ce 11 Area
1.15
(0.10)
1.18
(0.08)
1.33
(0.10)
Nucleus Length
0.94
(0.08)
0.76
(0.13)
0.90
(0.10)
Nucleus Width
0.97
(0.09)
0.96
(0.09)
0.97
(0.13)
Nucleus Area
0.95
(0.12)
0.73
(0.11)
0.86
(0.16)
NDR*
0.86
(0.28)
0.93
(0.24)
0.88
(0.16)
N r
15
7
15
+ Standard Deviation
* Nuclear Displacement Ratio (1 = no lateral displacement)
(0 = lateral displacement to the cell margin)

165
Table 19. Classification sumnary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with macrogametocytes . Discriminant scores were
classified with a discriminant function derived
from the calibration data set summarized in
Table 18.
Classified Into Species
Spec ies
Ch u c k a r
Pheasant
Turkey
Other
Tota 1
Chucka r
11 (73.3)+
0 (0.0)
3 (20.0)
1
(6.7)
15
(100)
Pheasant
3 (42.9)
0 (0.0)
3 (42.9)
1
(14.3)
7
(100)
Turkey
3 (20.0)
0 (0.0)
10 (66.7)
2
(13.3)
15
(100)
Total
17 (46.0)
0 (0.0)
16 (43.2)
4
(10.8)
37
(100)
Priors*
0.4054
0.1S92
0.4054
+ Standard Deviation
* Prior probability of being assigned to that class

166
Table 20. Classification sumnaty of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with macrogarnetocy tes . The discriminant
function derived from data in Table 18 was used
to classify the discriminant scores.
Classified Into Species
Species
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
1 (25.0)+
0 (0.0)
3 (75.0)
0
(0.0)
4
(100)
Pheasant
2 (66.7)
0 (0.0)
1 (33.3)
0
(0.0)
3
(100)
Turkey
1 (25.0)
0 (0.0)
2 (50.0)
1
(25.0)
4
(100)
Total
4 (36.4)
0 (0.0)
6 (54.6)
1
(9.1)
11
(100)
Priors*
0.4054
0.1892
0.4054
+ Percentage of total
* Prior probability of being assigned to that class

167
Table 21. Average adjusted measurements of host cells
infected with microgametocytes. Each variable
was divided by the average value of the
same variable from uninfected cells of the same
spec i es.
Variable Chuckar Pheasant Turkey
Ce 11 Length
1.17
(0.09)+
1.03
(0.07)
1.11
(0.06)
Cell Width
1 .02
(0.08)
1.17
(0.08)
I . 1 1
(0.09)
Ce11 Area
1.18
(0.10)
1.21
(0.08)
1.27
(0.11)
Nucleus Length
0.92
(0.11)
0.82
(0.11)
0.99
(0.11)
Nucleus Width
0.84
(0.11)
0.98
(0.17)
0.95
(0.10)
Nucleus Area
0.77
(0.17)
0.31
(0.15)
0.94
(0.16)
NDR*
0.89
(0.15)
0.28
(0.21)
0.42
(0.26)
N =
15
IS
15
+ Standard Deviation
* Nuclear Displacement Ratio (I = no lateral displacement)
(0 = lateral displacement to cell margin)

163
Table 22. Classification summary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells
infected with microgametocytes. Discriminant
scores were classified with a discriminant
function derived from the calibration data
set in Tab 1e 21.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Tota 1
Chuckar
9 (60.0)+
0
(0.0)
3
(20.0)
3 (20.0)
15
(100)
Pheasant
0 (0.0)
7
(46.7)
6
(40.0)
2 (13.3)
15
(100)
Turkey
3 (20.0)
2
(13.3)
9
(60.0)
1 (6.7)
15
(100)
Tota 1
12 (26.7)
9
(20.0)
13
(40.0)
6 (13.3)
4S
(100)
Priors*
0.3333
0
. 3333
0
. 3333
+ Standard Deviation
* Prior probability of being assigned to that class

169
Table 23. Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with microgametocytes. The discriminant
function derived from data in Table 21 was
used to classify the discriminant scores.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
2
(50.0)+
1
(25.0)
0
(0.0)
1 (25.0)
4
(100)
Pheasant
0
(0.0)
4
(100)
0
(0.0)
0 (0.0)
4
(100)
Turkey
1
(25.0)
1
(25.0)
1
(25.0)
1 (25.0)
4
(100)
Total
3
(25.0)
6
(50.0)
1
(8.3)
2 (16.7)
12
(100)
Priors*
0.
.3333
0.
,3333
0.
,3333
+ Standard Deviation
* Prior probability of being assigned to that class

Figure 59. Parasitemias per 10,000 red blood cells for turkeys,
(O O ), the Chuckar ( A A ) and the Ring-necked
Pheasant ( © © ) with experimental infections of
Ha einop roteus me 1 e a g r i d i s . Values for turkey were
calculated as the average parasitemia for the 2 infected
birds. All birds received the s ame n umb e r of
sporozoites.

WEEK
LOG|Q (Parasitemia + I)
o — ro oj
\L I

Figure
60.
Mac rósametocyte of Haemoproteus meleagridis
from an experimentally infected turkey.
Giemsa. Bar = 10 um.
Figure
61 .
Micrósametocyte of Haemoproteus meleagridis
from an experimentally infected turkey.
Giemsa. Bar = 10 um.
Figure
62.
Macrogametocyte of Haemoproteus meleagridis
from an experimentally infected C'nuckar.
Giemsa. Bar = 10 um.
Figure
6 3 .
Microsametocvte of Haemoproteus meleagridis
from an experimentally infected Chuckar.
Giemsa. Bar = 10 um.
Figure
64.
Macrogametoevte of Haemoproteus meleagridis
from an experimentally infected Ring-necked
Pheasant. Giemsa. Bar = 10 um.
Figure 65. Microgametocyte of Haemop roteus me 1e ag ridis
from an expe r i men t a 1 Ty infected Ring-necked
Pheasant. Gienisa. Bar = 10 um.

173
%
•• ^
•f
^60
• «
4. Ik.
O» ^
1 # 1
, »f «7
61
*
62
J
• i #l
J3
% <•"
*
/ ’#
â–  64 1
Jl
^ H; T 9
B ^65

174
macrogainetocytes, the endoplasmic reticulum contained a
moderately dense, amorphous material (Figure 68). The
endoplasmic reticulum was occasionally continuous with the
innermost, osmiophilic layer of the pellicle (Figure
68). Ma c r ogame t oc y t e s also contained membrane-bound,
osmiophilic bodies (Figure 67).
Gametocytes contained a single cytostome,
surrounded by 2 electron dense rings (Figure 69). Food
vacuoles containing host cell cytoplasm and limited by the
2 outer membranes of the pellicle formed at the inner
surface of the cytostome, between the 2 electron dense
rings. Other food vacuoles in the macrogametocyte were
bound by a single unit membrane and contained
osmiophilic masses of pigment (Figure 69).
Gametogenes i s
In the earliest stages of gametogenes i s, gametocytes
rounded-up within their host cells (Figures 70, 71,
74). The outer unit membrane of the 3-layered pellicle
separated from the parasite and formed membranous vesicles,
coils and whorls or remained free as discrete fragments in
the red blood cell cytoplasm (Figures 70, 72, 73).
Dense granules, approximately the same size and electron
density as free ribosomes, occasionally lined the inner
side of the membrane in a single layer (Figure 73). The
thickened, osmiophilic inner layer of the pellicle of

175
mi crogametocytes developed breaks and discontinuities
throughout its surface so that the middle layer of the
pellicle became the new, outer limiting membrane of the
parasite (Figures 71, 72). By contrast, the inner layer of
macrogametocytes remained intact (Figure 74).
Macrogametocyte nuclei were elongate and extended
across the diameter of the parasites. A prominent
nucleolus was present at 1 pole. A kinetic center composed
of a mass of amorphous, electron-dense material was
adjacent to the nuclear envelope at the other pole (Figure
74).
Within 3 minutes after gametogenesis began,
macrogametocytes and mi erógametocytes were free of their
host cells. Both were bound by the middle layer of the
original 3-layered pellicle. The pale, electron lucent
remnants of ruptured host cell nuclei as well as
remnants of host cell membranes were often adjacent to
or surrounded free gametocytes (Figure 77). Both maturing
macrogametes and exf1 age 1 I ating mi erógametocytes contained
a single cytostome (Figures 76, 78).
Maturing macrogametes were packed densely with
ribosomes and contained numerous mitochondria and an
extensive network of smooth endoplasmic reticulum
(Figure 75). They contained an elongate nucleus with a
prominent nucleolus that extended across the center of the
macrogametocyte (Figure 75).

176
Mi crogametocyt e s contained a large, diffuse nucleus
(Figure 77). Dense aggregates of electron dense
material with embedded microtubules were occasionally
located adjacent to the nuclear envelope (Figure 77).
Axonemes in various stages of assembly were scattered in
the cytoplasm of exf1 age 11 at ing microgametocytes and often
extended from the nucleus to the outer limiting membrane of
the parasite. In cross section, they consisted of 9
peripheral doublets of microtubules that surrounded 2
central tubules (Figures 77, 78). In longitudinal section,
the central tubules had regular striations along their
length (Figure 79). As development progressed, axonemes
budded from the outer surface of the parasite, between
interrupted portions of the osmiophilic inner layer of the
pellicle (Figure 80). Portions of the mi crogametocyte
nucleus were occasionally drawn to the base of flagellar
buds (Figures 80, 81). Exf1age11 ated microgametes
contained a single axoneme and a membrane-bound nucleus
(Figure 82).
Oocys ts
Three-day-old oocysts. Three-day-old oocysts were
subspherical in shape and surrounded by a thick, amorphous
wall (Figure 83). Oocysts were located under the basement
membrane of the midgut. The 2 structures had the same

177
electron density and were often indistinguishable in areas
where the basement membrane was stretched tightly over the
oocyst wall (Figure 83).
The pellicle surrounding the 3-day-old sporo'olast was
imme diately interior to the oocyst wall. It was
composed of a single unit membrane, underlaid in many
places with dark, osmiophilic thickenings (Figure 83).
Prominent, membrane-bound, lipid-like inclusions were
clustered together near the center of 3-day-old oocysts
(Figure 83). Surrounding these and scattered throughout
the cytoplasm were numerous mitochondria with tubular
cristae (Figure 83). Several irregularly shaped nuclei
with prominent nucleoli were present around the
periphery of the parasites. The nucleoplasm had a density
similar to the cytoplasm and was difficult to discern in
areas where the nuclear envelope was indistinct (Figure
83). Spindle fibers and kinetic centers were not observed.
Six-day-old oocysts. 3y 6 days, oocysts ranged in
development from immature forms, resembling 3-day-old
oocysts, to mature oocysts that had ruptured and
released their sporozoites. Oocysts that were more mature
than the 3-day-old forms had more space between the
sporoblast body and the oocyst wall. Numerous budding
sporozoites developed around the periphery of the
sporoblast. The buds originated under the osmiophilic

178
thickenings beneath the pellicle (Figure 84). As the buds
became more elongate, apical organelles including polar
rings, electron dense rhoptries and subpe1 1 i cu 1 ar
microtubules differentiated from the sporoblast body
(Figure 8S). One budding sporozoite had at least 18
subpe11 i cu 1 ar microtubules in cross section (Figure
85). The pellicle of developing sporozoites consisted
of an outer unit membrane that was underlaid by 2 unit
membranes in close apposition to one another (Figure 85).
Large nuclei were located at the periphery of the
sporoblast body, underneath the budding sporozoites.
Remnants of the apical complex of the ookinete, consisting
of an electron dense canopy and supporting microtubules
were present within the sporoblast (Figure 84). Large
lipid-like vesicles were clustered near the center of
the sporoblast body (Figure 84).
As sporozoites completed budding, the sporoblast body
became progressively smaller, leaving a residual mass
composed of large, lipid-like vesicles and amorphous
cytoplasm (Figure 86). At maturity, oocysts ruptured,
liberating sporozoites into the haemocoel of the
insect. Ruptured oocysts contained the degenerating
remnants of the residual body (Figure 87).

179
Mesaloschizont s
Mature, 19-day-old mega 1osch¡zonts were extracellular
and surrounded by a thick, laminated cyst wall composed o£
electron dense, amorphous material (Figure 88). Host
tissue adjacent to megaloschizonts was necrotic and
infiltrated with phagocytic cells (Figure 88). The region
between healthy muscle tissue and megaloschizonts contained
large quantities of amorphous, granular material with
the same electron density as the cyst wall (Figure 88),
membranous vesicles, free mitochondria with swollen and
ruptured cristae and scattered fragments of myofibrils.
Many mitochondria were opaque and electron dense. The
amo rphous, granular material exterior to the
megaloschizonts appeared to be deposited in layers onto the
outer surface of the cyst wall (Figure 88). Phagocytic
cells adjacent to ruptured megaloschizonts were often
packed with merozoites.
Megaloschizonts contained merozoites that developed as
buds from discrete cytomeres (Figure 89). The
merozoites were bound by a single unit membrane that was
underlaid by an interrupted intra-membranous layer composed
of 2 unit membranes in close apposition to one another
(Figure 89). Merozoites had 3 anterior polar rings, a pair
of electron-dense rhoptries, micronemes, a nucleus bound by
2 unit membranes and a single mitochondrion with tubular
cristae (Figures 89, 90, 91). All merozoites also had a

180
large, electron lucent, membrane-bound vacuole that
occupied from 1/4 to 1/3 of the total cytoplasmic volume
(Figures 90, 91).

Figure 66. Circulating microgametocyte. The parasite
is bound by a 3-layered pellicle with a
thickened, osmiophilic inner layer (arrow).
Other organelles include a nucleus (N),
mitochondria (M) with tubular cristae and food
vacuoles (Fv) that contain pigment. X 35,500.

182
66

Figure 67. Circulating macrogametocyte. The parasite
is bound by a 3-layered pellicle (arrow) and
has a branched nucleus (N) with a prominent
nucleolus (Nu), mitochondria (M) with
tubular cristae and osmiophilic bodies (double
arrow). The endoplasmic reticulum (Er)
contains a moderately dense, amorphous
material. X 46,400.
Figure 68. Higher magnification of Figure 67. The
endoplasmic reticulum (Er) is continuous
with the inner, osmiophilic layer of the
pellicle (arrow). X 72,000.

184

Figure 69. Circulating macrogametocyte. The gametocyte
has a cytostome (arrow) with an associated food
vacuole (Fv) derived from an indentation of the
2 outer layers of the pellicle (arrow). Older
food vacuoles (double arrow) are bound by a
single unit membrane and contain diffuse,
electron dense masses of pigment (?). X 37,000.

186

Figure 70. Maturing macrogamete. The outer layer of
the pellicle has separated from the parasite
and formed membranous coils (arrows).
Mitochondria (M) and endoplasmic reticulum
filled with amorphous material are scattered in
the cytoplasm. X 58,000.

188

Figure 71.
Exf1 age 11 ating mi erógametocyte . At the same
time the outer layer of the pellicle separates
from the parasite (double arrow), breaks appear
in the inner osmiophilic layer (arrow). X
46,400.
Figure 72. Higher magnification of Figure 71. X 72,000.

190

Figure 7 3.
Figure 74.
Maturing macrogame te . The outer layer of
the pellicle has separated from the gamete and
formed membranous whorls that are lined on
their exterior by dense granules that are
the same size and electron density as free
ribosomes (arrow). X 66,700.
Maturing macrogamete. The nucleus (N) is
elongate and extends across the diameter of the
gametocyte. A nucleolus (Nu) is present at one
end of the nucleus and an electron dense
mass (large arrow) with faint, embedded
microtubules (small arrow) is located at
the opposite end. X 58,000.

192

Figure 75.
Figure 76.
Extracellular maturing rnacrogamete. The gamete
has a pellicle composed of 2 continuous layers
(arrow), an elongate nucleus (N) with a
nucleolus (Nu), mitochondria (M) and an
extensive network of endoplasmic reticulum
(small arrows). X 34,800.
A higher magnification of a portion of
Figure 75. The macrogamete has a cytostome
(large arrow) that is surrounded by 2 electron
dense thickenings. The pellicle is composed of
2 uninterrupted unit membranes (small
arrows). X 74,750.

194

Figure 77. Extracellular exf1 age 11 at ing mi crogametocyte.
The remnants of the host, red blood cell
nucleus are adjacent to the gametocyte
(large arrow). The parasite has a large
diffuse nucleus (N) with aggregates of electron
dense material with embedded microtubules
adjacent to the nuclear envelope (small
arrows). Axonemes (A) and mitochondria (M) are
scattered in the cytoplasm. X 39,000.

196

Figure 78. Extracellular exf1 age 1lating mi erógametocyte.
Axonemes (A) in various stages of assembly are
associated with arm-like extensions (double
arrows) of the gametocyte nucleus (N) . The
gametocyte has a cytostome (large arrow),
mitochondria (M) and a 2-layered pellicle with
a discontinuous inner layer. X 46,400.
Figure 79. Extracellular exf1 age 11 ating mi erógametocyte .
The central microtubules (arrows) of the
transversely sectioned axoneme have periodic
striations along their length. X 66,700.

198

Figure 80. Extracellular exf1 age 11 at in^ microgametocyte.
Microgametes (large arrows) that contain a
single axoneme bud from the outer surface of
the mi erógame tocyte, between disruptions (small
arrows) in the inner layer of the pellicle.
The mi erógametocyte nucleus (N) is stretched to
the base of the budding microgametes. X
58,000.
Figure 81. Extracellular exf1 age 11 at ing mi erógametocyte.
The microgametocyte nucleus is stretched to the
base of the budding microgamete (arrow). X
49,200.
Figure 82. Cross sections of microgametes. Each
microgamete contains a single axoneme (A) and a
mass of nuclear material (N). X 78,000.

200

Figure 83. Three-day-o1d oocyst {coin a specimen of
Cu 1 icoi des edeni . The oocyst is surrounded by
a thick, amorphous wall (large arrows) and is
located beneath the basement membrane (3m) of
the midgut. The pellicle of the parasite
is underlaid by osmiophilic thickenings (small
arrows). Nuclei (N) with prominent nucleoli,
large membrane-bound, lipid-like inclusions
(L) and mitochondria (M) with tubular cristae
are present. X 26,100.

202

Figure 84. Six-day-old oocyst from a specimen of
Cu 1 ic o id e s e d e n i . Budding sporozoites
(large arrows) are spaced around the periphery
of the sporoblast body. Large nuclei (N)
are located are located at the periphery of the
sporoblast body, beneath the budding
sporozoites. The remnants of the electron
dense apical complex of the ookinete (Ac),
as well as large, lipid-like inclusions (L),
are near the center of the oocyst. X 26,100.
Figure 85. Six-day-old oocyst from a specimen of
Cu 1icoi des edeni. Budding sporozoites contain
e1ectron dense rhoptries (R), a nucleus (N), a
polar ring (small arrow) and subpe1 1 i cu 1 ar
microtubules (large arrows). The outer
layer of the pellicle that surrounds the
sporozoites is underlaid by 2 unit membranes in
close apposition to one another (double
ar row). X 78,300.

204

Figure 86. Six-day-old oocyst from a specimen of
Cu 1icoi des edeni. Budding sporozoites (arrows)
contain electron dense rhoptries (R). As they
complete their maturation, a small residual
body containing large, lipid-like vesicles (L)
rema ins. X 34,800.

206

Figure 87. Six-day-old oocyst from a specimen of
Cu 1 i coi de s edeni. The oocyst has ruptured,
releasing the mature sporozoites. A
degenerating residual body (R3) containing
lipid-like vesicles (L) and electron dense
masses (arrows) remains. X 26,100.

208

Figure 88. Mega 1oschizont and associated phagocytic cells
(PC) from pectoral muscle of a high dose
bird that died spontaneously at 20 days
post-infection. The wall (W) of the
megaloschizont is thick and laminated.
Large quantities of amorphous, granular
material are exterior to the wall. A mature
merozoite (Me) is located within the
interior of the cyst. X 26,100.

210

Figure 89. A cytomere with budding merozoites (arrows). A
nucleus (N) in the process o£ division is
constricted into 2 lobes. Budding
merozoites contain a large, electron-lucent
vacuole (V), rhoptries (R) and a mitochondrion
(M). Merozoites have a pellicle composed of an
outer unit membrane and a discontinuous
inner layer consisting of 2 unit membranes
in close apposition to one another (small
arrows). X 58,000.
Figure 90. Mature merozoite. The merozoite has a
large, e1ectron-1ucent vacuole (V), paired
rhoptries (R) and 3 anterior polar rings
(arrows). X 58,000.
Figure 9!. Mature merozoite. The merozoite has a
large, e 1 ectron-1ucent vacuole (V), paired
rhoptries (R) and 3 anterior polar rings
(arrows). X 58,000.

212

DISCUSSION
5pi zooti o 1ogy
Vectors
The functional vectors of any arthropod-borne
infection must be present at a density sufficient to
maintain transmission of the organism, must use the host
species as a regular source of blood meals and must be
susceptible to development of the parasite (Sates, 19+9;
Russell, 1 959). Of the 29 species of Cu I iciodes that were
captured in New Jersey light traps at Paynes Prairie and
Fisheating Creek, individuals of only 12 species took blood
meals from turkeys exposed in Bennett traps. Biting
collections
o f
7 of these 12 specie
s ,
i . e
arboricola.
C.
crepuscularis, C. gutt
i p e
n n i s
haema topot u s ,
c.
hinmani, C. paraensis and
a
scan
have been made from both birds and mammals (Blanton and
Wirth, 1979). Engorged specimens of O baueri have been
collected previously from mammals and specimens of C.
ou s airani have been collected from birds (Blanton and
Wirth, 1979). Biting records for edeni, C. knowlton i
and O nanus have not been reported previously.
213

214
The total numbers of edeni and hinmani collected
in Bennett traps at Paynes Prairie and Fisheating Creek
were considerably larger than collections of any of the
other 10 species, '«/hile individuals of both species
were capable of supporting complete development of the
sporogonic stages of me 1 e a g r i d i s , the greater
susceptibility of specimens of C^ edeni and their
preponderance in Bennett trap collections throughout the
year indicate that this species is the primary vector of H.
me 1e a g ridis in Florida.
Specimens of arbor icoI a made up approximately 8% of
the collections from Bennett and New Jersey traps at Paynes
Prairie. This species may be conrnon enough to play a minor
role in the transmission of me 1eag rid i s in northern
Florida. However, total numbers captured at Fisheating
Creek were insignificant when compared to the large numbers
of C^ eden i and C^ h i nman i that were taken at the same
time.
Cu 1 icoi des knowltoni is found rarely north of central
Florida and specimens were not captured at Paynes Prairie.
Individuals of this species composed a small fraction
(1.4%) of the Bennett Trap catch at Fisheating Creek,
but were taken more commonly in New Jersey light traps.
This discrepancy suggests that individuals of this species
use other hosts as a blood source and may not play a
significant role in the epizootiology of H_^

215
meleagridis . Cu 1 i c i od e s crepuscularis is related
closely to knowl ton i and replaces it in northern Florida
and throughout Nortii America (Blanton and Wirth, 1979). It
is the only species of Cu I ico i des reported from Florida
that has been implicated previously as a vector of avian
haemoproteids (Bennett and Fallís, 1960). However,
specimens of crepuscularis were collected rarely in
Bennett Traps and light traps at Paynes Prairie and it
is unlikely that this species plays an important role in
the natural transmission of H_^ meleagridis in Florida.
Specimens of C_^ h a ema t o po t u s were captured too
infrequently at Paynes Prairie and Fisheating Creek to
implicate this species as a vector of meleagridis at
either site. However, it is related closely to C^ edeni
(31anton and Wirth, 1979). Since the distribution of C.
eden i is limited to Florida and the Bahamas (Blanton and
Wirth, 1979), C^ haematopo tus may be an important vector in
other parts of North America where the prevalence of H.
me 1eagridis is high.
Sporogonic Stages and Experimental Transmission
The size and morphology of the ookinetes, oocysts and
sporozoites of me 1eag rid i s , as well as the relatively
short sporogonic cycle of 6 - 7 days, resemble the findings
for other species of Haemop rot eu s known to develop in
ceratopogonids (Fallís and Wood, 1 957; Khan and Fa 11 is,

216
1971; Fallís and Bennett, 1961; Miltgen et al., 1981). The
relatively low (50%) infectivity of salivary gland
sporozoites ft om specimens of eden i , hinman i and
C. arboricola as well as the lack of infectivity of
sporozoites from the single specimen of haematopotus,
may be related to the problems inherent in handling sticky
salivary glands that are only 100 - 200 um long.
Another possibility is that the sporozoites observed in
these wild-caught specimens of Cu 1 icoides belonged to
another species of haemosporid i an . Bennett and Coombs
(1975) reported a sporozoite prevalence of 13.6% in 184
individuals of sti lobezziodes f r om insular
Newfoundland. However, only 1 individual (0.5%) out of 210
unengorged, wild-caught specimens of ed e ni from
Paynes Prairie was positive for salivary gland sporozoites
by dissection, in spite of the fact that collections
were made during periods of active transmission of H.
me 1 eagridis. Natural infections were not detected in
salivary gland dissections of unengorged specimens of C.
arbor icola or C_^ h i nma n i . Attempts to transmit H.
me 1eag ridis with individual Cu 1 icoi des and with pools of
several specimens that had been ground in a suitable
carrier were more successful.
The 17 day prepatent period of successful experimental
infections was less than the 28 day period that Greiner and
Forrester (1980) estimated from observations of sentinel

217
turkeys, exposed in Wild Turkey habitat. It is similar to
the shorter prepatent periods of other haemoprote i d s
transmitted by species of Cu 1 i co i des ; 14 - 21 days for
H. ne11ionis, 14 days for mansoni and 11-14 days for H.
ve 1 an s (Fall is and Wood, 1957; Fall is and Bennett, 1960;
Khan and Fallís, 1971).
Activity Cycles
Most species of Cu 1 icoides have crepuscular and/or
nocturnal peaks in their daily activity cycles that may
play an important role in their contact with potential
hosts (Kettle, 1965, 1977; Barnard and Jones, 1980).
Kettle (1968b) noted that this cycle is probably an
endogenous circadian rhythm, regulated by changes in light
intensity and modified by local meteorological
conditions such as wind velocity and temperature. A number
of sampling techniques have been used to measure diel
changes in activity, including truck-mounted
interception traps (3 id1ingmeyer, 1961), suction traps
(Service, 1971), biting collections (Kettle, 1968a) and
paddle traps (Nathan, 1981). Studies that have employed
more than 1 sampling method have generally found them to be
in close agreement in detecting major peaks of activity
during the 24-hour cycle (Nathan, 1981; Service, 1971).

218
The large, evening Bennett trap catches of
specimens of C^_ eden i , CC hinmani, C. arboricol a and C.
knowlton i suggest that individuals of these species are
primarily crepuscular with major peaks in biting
activity near sunset. Since sampling with the Bennett
traps was restricted primarily to the 2 hours preceding and
following sunset, peaks of activity at other times of
the day would have been missed. However, the limited
amount of Bennett trapping that was done at night, at dawn
and during the day did not detect any major periods of
activity.
Peak evening collections of specimens of O hinmani
were consistently and significantly earlier than
collections of the other species (p < 0.05). Differences
among mean capture times for specimens of eden i , C.
arboricol a and knowltoni were not consistent from
quarter to quarter or site to site. The large amount of
overlap among capture times for individuals of these
species and the correspondingly large standard deviations
for mean capture times at each quarter and site, suggests
that the differences are not biologically significant.
The Bennett trap catches and the CC>2_baited suction
trap catches of specimens of edeni and hinmani
were similar at Fisheating Creek. Individuals of both
species had peaks in their biting/host-seeking activity in
the forest canopy during the 2-hour sampling period that

219
included sunset. Suction trap catches of specimens of
C. e d e n i and h i n in a n i fell rapidly in numbers
following twilight and remained low until after sunrise,
when catches of individual hinmani had a second, smaller
increase in numbers. Individuals of both species were
active at lower levels during the day on 1 or more of the 3
collection dates .
Few specimens of eden i and no specimens of C.
h'nmani were captured in the CC>2-baited suction traps
operated near the ground. However, the numbers of
individual eden i observed on the exposed sentinel
turkeys during the day, as well as the high prevalence
of transmission of me 1 e a g r i d i s to sentinel birds
caged on the ground, indicates that ground-level biting
activity occurred. The ground traps may have been located
in a poor position or the technique may not have been
sensitive enough to detect low numbers of Cu 1 icoi des. Snow
(1955) noted that biting activity of individuals of C.
paraens i s , C. s pinos u s and borinquen i spread upward
along the main trunk of the tree and then outward into the
canopy during the day. Since the ground traps were not
positioned near tree trunks, they would not have
detected similar movements by specimens of eden i or
C. hinmaniâ–  Other studies have found distinct differences
in the vertical distribution of Cu 1 icoides species in
forest habitat (Snow, 1955; Bennett, I960; Service,

220
1971; Tanner and Turner, 1974). In general, ornithophi 1 i c
species were active in the forest canopy, where avian hosts
are presumably more available. Tanner and Turner (1974)
suggested that species of Cu 1 ic oid e s may occupy a
particular vertical stratum and search for a suitable host
within that stratum, regardless of whether it is a bird or
a mamna 1 .
Wild Turkeys normally roost at night in the middle
levels of the forest canopy (Schorger, 1966) and typically
fly to a suitable branch at sunset when individuals of
C. eden i , C. arbor ico1 a and knowlton i are becoming
active. The significantly higher levels of transmission of
H. me 1eagridis that occurred to sentinel birds exposed
in the canopy at Paynes Prairie and Fisheating Creek
(Forrester, unpublished data) supports the evidence that
individuals of 1 or more of these species serve as vectors
of the parasite. Since total numbers of h i nman i
captured in Bennett traps and suction traps were
greatest before Wild Turkeys roost for the night and, since
this species was never captured in Bennett traps or suction
traps operated near the ground, it may play only a minor
role in the ep i zoo t i o I ogy of me 1 eagr i d i s .
Transmission and Vector Abundance
Since the discovery of avian haemosporidians by
Danilewsky ( 1 889 ), few epizoot i o 1ogica 1 studies have

221
been made o£ any of the avian species of P1 a smod i um,
Leucocytozoon o r Haeinopcoteus. Those that have been
conducted have relied primarily on repeated surveys of the
host population by blood smears to monitor the
prevalence of the parasites under study (Herman, 1938;
Herman et al., 1954; Janovy, 1966; Bennett and Fallís,
1960; K1ei and DeGiusti, 1975; Greiner, 1975). While this
procedure has the advantage of monitoring changes directly
in the host population, it is limited by the
difficulties and biases inherent in capturing wild birds
and diagnosing latent and low intensities of infection. In
addition, the precise time when transmission of a blood
parasite begins and its relationship to vector populations
may be difficult to establish.
Chernin (1952) was one of the first to overcome these
problems by using domestic sentinel ducks to monitor
natural transmission of Leucocytozoon simondiâ–  Later
workers, including Fa 11 is and Wood (1957), Fa 11 is and
3ennett (1966) and Herman and Bennett (1976), refined these
techniques in more extensive studies of L^ simondi and
applied them to the study of Haemoproteus ne11ionis. In
conjuction with surveys of wild hosts, the use of sentinel
birds provides a powerful tool for studying the dynamics of
natural transmission. By decreasing the length of time
sentinels are exposed, it is possible to measure the onset
of transmission to within days. Sentinel studies are

222
limited, though, by the availability of a suitable domestic
host for the blood parasite under investigation. As a
result, this technique has been limited primarily to the
study of Leucocytozoon and Haemoproteus in ducks.
Forrester (unpublished data) and Akey (1981), at
Fisheating Creek, were the first to use domestic, sentinel
turkeys to monitor the natural transmission of Hâ– 
me Ieagridis. Earlier surveys in this area by Forrester et
al. (1974) established that me 1eagr i d i s had a prevalence
that ranged from 70 - 100% in Wild Turkeys older than 5
months. The increasing prevalence of me 1 eagrid i s in
juvenile birds collected at monthly intervals up to 1 year
after hatching suggested that transmission of the parasite
occurred throughout the year (Forrester et al., 1974). In
concurrence with this, Forrester (unpublished) found
that transmission occurred to 50- 100% of the sentinel
poults that he exposed for 12, 2-week intervals between
May, 1978 and May 1979. Akey (1981) had similar results
during a shorter study between May and August, 1980, in the
same area. Both Forrester (unpublished) and Akey
(1981) noted that transmission of me 1eag rid i s
occurred at higher prevalences among birds caged in the
canopy, than to those caged on the ground. Results of the
present study confirm these previous observations at
Fisheating Creek and extend the observations from the warm,

223
subtropical climate of southern Florida to the more
temperate climate of northern Florida.
The high prevalence of transmission of roe 1eagrid i s
to sentinel turkeys at Fisheating Creek was correlated with
the large number of specimens of edeni that were
captured in light traps and 3ennett traps throughout the
year. The repeated isolations of me 1eagridis from pools
of wild-captured, unengorged specimens of O edeni provide
further evidence of the importance of this species as a
major vector of meleagridis. The conspicuous
absence of individuals of hinmani, C. arbor ico1 a and C.
knowlton i at times when transmission was high, their lower
biting
rates and the fa i
lure
to obtai
n wild i s o1 a t
i ons
of the
parasite fr oin i n
d i v j
dual s
o f
these speci
e s ,
indicates that they may
play
only
mi
nor roles in
the
epizootiology of me 1eagrid i s .
The low numbers of specimens of eden i that were
captured during abnormally cool, wet weather in February
and March, 1 983 and November, 1984, and the absence of
detectable levels of transmission of me 1eagridi s may be
related. Greiner (pers. c onm.) also failed to detect
transmission of HL meleagridis to sentinel turkeys in
the same area during abnormally wet weather in March and
April, 1978. It is unclear whether the population of
adult Cu 1 icoi des was inactive because of cool temperatures

224
or whether a significant reduction in the population
size had occurred.
Fisheating Creek is subject to frequent flooding
and wide fluctuations in water levels and flow (U.S.
Geological Survey, Water Resources Division, Orlando,
Florida). During the drier months of the year (November -
March), flow may be reduced to a trickle and the stream can
become a series of shallow pools and ponds. During
heavy rainstorms, stream levels may rise as much as 7 feet
in 24 hours. Because the topography in the area is very
flat, ranging from 30 to 55 feet above sea level, the
creek swamp may be flooded up to 1 km or more on either
side of the stream channel. Since individuals of many
species of Cu I i c o i d e s , including C^ eden i , breed in
moist soil or mud at the edges of permanent or
semi-permanent bodies of water (Kettle, 1977),
extensive flooding may erode breeding sites and wash the
immature stages away. Corresponding reductions in adult
populations may reduce the numbers of potential vectors to
below the critical density needed to maintain continuous
transmission of the parasite.
Patterns of transmission and vector abundance at
Paynes Prairie were similar to those at Fisheating Creek.
The year-round activity of individuals of Ch edeni, the
isolation of me 1eagridis from specimens of naturally
infected C^ edeni, and the conspicuous absence of

225
individuals of
C.
hinmani
and
C. arboricola
a t
times
when natural transmission
of
H. me 1e a g ridis
wa s
h i gh
indicates that
c.
eden i is
the
primary vector
a t
Paynes
Prairie as well. However, several important differences
between the 2 study areas in northern and southern Florida
were evident. Transmission of me 1 eagr i d i s occurred at a
lower prevalence at Paynes Prairie than it did at
Fisheating Creek and was not continuous throughout the year
(Figures 13, 14, 18). It ceased during the cooler
winter months between January and April, when average
monthly temperatures were below 60o ancj wa s often
interrupted by periods of 2 or 4 weeks at each of the 2
study sites during the warmer months of the year. In
addition, Site A and Site 3 at Paynes Prairie had
significant differences in the time of onset of
transmission in the spring, the time when major and
minor peaks occurred and the time when transmission ceased
in the winter (Figures 13, 14). Unlike the collections at
Fisheating Creek, collections of specimens of O edeni were
not closely correlated with peaks in transmission of H.
me I eagridis at Paynes Prairie (Figures 13, 14).
Censuses of the «/ild Turkey population in Florida over
the past 30 years have shown dramatic increases throughout
the state (Powell, 1967). Surveys have consistently shown
that the population in Alachua County, where Paynes Prairie
is located, is less than 1/10 as large as it is in

226
Glades County, at Fisheating Creek (Powell, 1967; L.E.
Williams, pers. comm.). In addition, Wild Turkeys from
northern Florida have a lower prevalence of me 1eagridis
(Forrester et al., 1974). Since many species of Cu 1 icoi des
have flight ranges of less than I km (Kettle, 1977),
transmission of me 1eagridis at Paynes Prairie may be
very local and dependent on the daily movements of Wild
Turkeys and their selection of roosting sites for the
night. A number of studies have shown that Wild Turkeys
maintain home ranges of from 400 - 1500 hectares in the
southeastern U.S. and, depending on the season, often move
in regular patterns and reuse favorite roosting sites
(Schorger, 1966; Bowman et al., 1979). Since Bennett trap
collections and light trap collections of C^ edeni were of
comparable sizes at Sites A and B, the higher
prevalences of transmission that were observed at Site
B, their earlier peaks and their longer seasonal
duration may have been due to greater use of the area by
Wild Turkeys. Several authors have reported that Wild
Turkeys prefer open woodland, pastures and old fields
for foraging and nesting (Stoddard, 1963; Lewis, 1964).
Since Site 3 was located in an ecotone between an open
field and the deciduous forest, it may have been more
attractive to Wild Turkeys and received greater use. As a
result, potential vectors of me 1 eag r i d i s would be
more likely to become infected with the parasite. The

227
lower prevalence o£ sporozoites in naturally infected C.
eden i could be a reflection of the smaller Wild Turkey
population at Paynes Prairie and the lower prevalence of
available gametocytes for potential vectors.
Studies of ? 1 a sinod i urn v i vax and P^ falciparum have
shown that development of the sporogonic stages in
mosquitoes is inhibited completely below temperatures of 15
- 20o q (Russell, 1959; Wernsdorfer, 1980). The
temperature dependent development of the sporogonic stages
is believed to be the major factor limiting the
temperate distribution of P^ v i vax and P^ falciparum to
areas within the 20o C summer isotherm (Wernsdorfer,
1980). Morii et al. (1965) observed a similar temperature
dependence in the development of the sporogonic stages
of caul 1 e r y i . Development of sporozoites in C.
ci rcumscr ip tus and arakawae was inhibited at
temperatures below 15 - 20o C. They suggested that
lower environmental temperatures may be an important factor
in the lower prevalences of caul 1e r yi in Japanese
poultry during the autumn.
Average monthly temperatures at Paynes Prairie and
Fisheating Creek are similar between May and October,
but range from 3 - 6o C cooler at Paynes Prairie between
November and April (Figure 17). The absence of
transmission of me 1eagr i d i s during February, March
and April, 1983 at Site A, when biting activity of Ch edeni

228
was still high, suggests that the inhibition of
sporogony by lower environmental temperatures may be 1
factor limiting the winter transmission of H.
me 1eag r i d i s at Paynes Prairie. During this study,
biting activity was never detected for any species of
Cu 1 icoid e s below ISo C. Inhibition of host-seeking
activity by cool evening temperatures between January
and March may also be a major factor in restricting the
winter transmission of me 1eagridis.
The ep i zoo t i o 1 og y of fh meleagridis in southern
Florida has many similarities to stable, holoendemic
malaria (Wernsdorfer, 1980; Gabaldon and Ulloa, 1980). At
any given time during the year, most of the Wild Turkey
population at Fisheating Creek is infected with H.
me 1eagrid i s and has circulating gametocytes, available for
potential vectors. The high prevalence of H â– 
me 1eagridis and the large, relatively stable vector ( =
C. edeni) population insure that a high rate of natural
transmission can occur throughout the year. In these
respects, the epizootiology of this parasite is
significantly different from other avian haemosporid i ans
that have been studied in temperate North America. In
temperate locations, populations of vectors as well as the
prevalence of avian haemosporidians are highly
seasonal. It has been documented that the spring
relapse of Haemop r o t eus , PI asmod i um and Leucocy tozoon i s

229
under hormonal control and occurs in conjuction with the
reproductive cycles of their avian hosts (Haberkorn, 1968;
Desser et al., 1968; Rogge, 1968; Applegate, 1970).
Transmission of the parasites occurs during a relatively
short period when recently hatched, susceptible young
are leaving their nests, when vector populations have
increased with the onset of warmer weather and when the
older adult population, with chronic, relapsing infections,
is available as a reservoir of infection (Janovy, 1966;
Fallís and 3ennett, 1966).
The more temperate climate at Paynes Prairie, the
smaller and more variable vector populations and the lower
prevalence of infection in the host population are similar
to conditions of hyperendemic, unstable malaria
(Wernsdorfer, 1980). Wide variations in the prevalence of
infection may occur, depending on environmental conditions,
host density and vector abundance and bionomics. However,
in contrast to the more unstable, epidemic transmission of
other avian blood parasites in northern North America,
transmission can remain high throughout most of the year.

230
Pathogen icity
Exoerythrocytic Development
The description of the schizonts of columbae by
A r a g a o (1908) was the first detailed account of the
exoerythrocytic development of any haemosporidian parasite
(Garnham, 1966). In the 77 years since then, little
more has been learned about the exoerythrocytic development
of avian haemop r o teids . Because experimental infections
have been difficult to perform for most species of
Haemoproteus, virtually nothing is known about their early
stages of exoerythrocytic development.
The results of this study demonstrate that H.
me 1eagridis undergoes at least 2 generations of schizogony
in skeletal and cardiac muscle of experimentally
infected turkeys. Following i n traperitonea 1 inoculation,
sporozoites probably gained access to the circulation
via the lymphatic system and became localized in the
rich capillary beds of the skeletal muscle. It is unclear
whether they began their development within capillary
endothelial cells, satellite cells or myofibroblasts. The
early schizogonic stages of Sarcocy s tis may occur in all 3e
cell types, as well as macrophages, but are most conation in
endothelial cells (Entzeroth, 1983; Dubey et al., 1983;
Cawthorn et al., 1983; Leek et al., 1977). The early
schizogonic stages of other cyst-forming sporozoans,

231
e . g. Toxoplasma and Hammondia, develop in other organs
or tissues (Frenkel and Dubey, 1975; Frenkel, 1973).
The single, 3-day-old schizont that was found appeared
to be within the capillary endothelium (Figure 42).
However, the presence of disrupted muscle fibers in the
infected bird and the development of 5-day-old schizonts
both within and between muscle fibers suggests that
development may occur in other locations as well
(Figures 43, 44, 53).
Studies of other avian haemoproteids have reported
development of the mature exoerythrocytic stages in
capillary endothelial cells of lung, liver, spleen, heart,
kidney and cecum (Khan and Fallis, 1969; Ahmed and
Mohammed, 1977; Bradbury and Gallucci, 1971; Sibley and
Werner, 1984; 3aker, 1966a; O'Roke, 1930; Greiner, 1971).
Farmer (1965 ) found large mega 1 oschizonts within the
gizzard muscles of Rock Doves with natural infections of H.
sacharovi. He suggested that they may be the
exoerythrocytic stages of H^ sacharovi , but was unable
to transmit the infections with transplants and injections
of infected gizzards. Similar forms have not been found in
Mourning Doves infected with HL sacharovi (Farmer, 1965;
Greiner, 1971) and most workers have ignored the
possibility that they may be part of a haemoproteid life
cycle. Mi 1tgen et al. (1931) reported the development
of megaloschizonts within muscle tissue of

232
8 i ossum-Headed Parakeets, naturally infected with H.
desser i. They noted that development appeared to be within
muscle fibers, rather than cells of the reticuloendothelial
system. In their study, as well as the one by Fanner
( 1965), the mega 1 os ch i zon t s were too large to precisely
determine the type of host cell.
Mature, 5-day-old, first-generation schizonts of H.
me 1eagridis contained elongate zoites that differed in size
and morphology from the pre-erythrocytic merozoites of
mature, 17-day-old, second-generation mega 1 osch i z on t s
(Figures 44, 50). They were approximately the same size as
first-generation hepatic schizonts of simo n di , but
lacked the hypertrophied host cell nucleus that is
characteristic of Leucoc y t ozoon infections (Huff,
1942). In addition, first-generation merozoites of
Leucocytozoon are oval to round in shape, rather than
elongate (Desser, 1967, 1974; Akiba et al., 1971; Huff,
1942). The first-generation schizonts of H^ me 1eagrid i s
were more similar in morphology to the first-generation
schizonts of Sarcocystis (Leek et al., 1977; Dubey et al.,
1983). The small size of schizonts found in the 8-day-old
infection suggests that the first-generation schizonts
ruptured between 5 and 8 days post-infection and
released zoites that initiated a second-generation of
schizogony.

233
Failure to detect developing schizonts at 11 days
post-infection may have been because the infected bird had
an infection of low intensity. The large difference in
size between the 8-day-old and the 14-day-old forms
makes it unlikely that a third generation of schizogony
could have taken place in other tissues.
The large size of the second-generation
illegal osch i zonts of me 1 e a a r i d i s is comparable to
mega 1oschizonts that have been found in birds with natural
infections of ga r nil ami , desser i and s acharov i
(Garnham, 1966; Miltgen et al., 1981; Farmer, 1965).
However, the mega 1 oschizonts of garnhami contained
cytomeres, separated from each other by distinct septa,
lacked a thick, hyaline wall and have not been found in
muscle tissue (Garnham, 1966). Miltgen et al. (1981)
described "pseudo-septa" in inmature megaloschizonts of H.
de s s e r i . The authors suggested, though, that these were
artifacts caused by contraction of the surrounding
muscle fibers during fixation of the tissue. The presence
of a hyaline cyst wall around megaloschizonts of H,
de s se ri, the absence of septa in mature forms and the
exclusive development in muscle tissue were very similar to
H. me 1eag ridis. The thick-walled, aseptate cysts
described by Farmer (1965) were also very similar to mature
megaloschizonts of me 1eagridis, but their development
appeared to be limited to gizzard muscle. Immature

234
mega 1 oschizonts were not found (Farmer, 1965). Since
sequential observations of the development of g a r n h ami,
H, de s s e r i and the forms described by Farmer (1965) have
not been made, the significance of cytomere development in
mega 1 oschizonts of me 1eagr i d i s cannot be determined.
The progressive formation of smaller and smaller cytomeres
is most similar to that described for mega 1oschizonts of
leucocytozoon (Huff, 1942; Khan and Fallís, 1970).
However, final merozoite formation in species of
Le ucocy tozoon occurs by fragmentation of the cytomere
cytoplasm rather than by bud formation (Desser, 1970).
Cytomere formation has not been confirmed in other
species of Haemoproteus. Aragao (1908) described the
segmentation of mu 11inuc1eated schizonts of Ih co1umbae in
lung tissue into numerous uninucleate masses that underwent
further nuclear division and growth. Wenyon (1926) termed
these masses cytomeres. Other studies have not observed
this process during the exoerythrocytic development of
Hâ–  co1umbae (Ahmed and Mohammed, 1977; Mohammed, 1965;
Bradbury and Gallucci, 1971). JJ1trastructura 1 studies
by Bradbury and Gallucci (1972) revealed that
mu 11inucleated masses of cytoplasm developed as clefts and
projections from the parent schizont. Since these did not
develop from discrete, uninucleate masses and did not
always detach from the parent body, Bradbury and
Gallucci (1972) used the term "pseudo-cytomere" as used by

235
Garnharn (1951) to describe them. Bray (1960) attempted to
resolve differences between the terms "cytomere" and
"pseudo-cytornere" and simplify terminology by redefining
cytomeres as "separate growing nucleated masses,
produced within a process of schizogony by the division of
a schizont.". He stated that the process "may produce
further cytomeres but produces uninucleated organisms as an
end point." Unfortunately, the redefinition has not
been widely accepted. Most authors have applied the
term "pseudo-cytomere" to descriptions of schizonts that
contain mu 11 i nuc1eated masses of protoplasm that remain
attached to the parent schizont (Bradbury and Gallucci,
1972; Sterling and DeGiusti, 1972). Since knowledge of the
entire developmental sequence is required to determine
whether a mu I t i nuc 1 ea t ed mass developed from a separate
uninucleate body or from a multinucleated mass that
detached late in development from the parent body, use
of either term may be misleading when the complete
morphogenesis is unknown.
During the development of second-generation
mega 1 oschizonts of meleagridis, cytomere formation
occurred between 8 and 14 days pos t-i n f ect i on . At the
present time it is not possible to determine whether the
cytomeres developed by segmentation of the mu I tinuc1eate
8-day-old schizonts into uninucleate masses or by
detachment of multinucleated masses from the parent body.

236
Since 14-day-old meg a 1 oschizonts contained numerous
discrete bodies that resembled the cytomeres described
in developing megaloschizonts of Leucocytozoon, the
terminology as defined by 3ray ( 1960) has been applied
to megaloschizonts of me 1eagridis.
The severe lameness and extensive myopathy that
occurred in infected birds during development of the small,
first-generation schizonts of me 1eagridis is unusual for
haemospor i d i an infections. The early, pre-erythrocytic
stages of most species of PI a smodium do not elicit any host
reaction and later exoetythrocytic schizonts appear to
be pathogenic only during unusual circumstances when the
parasite is exceptionally infectious to the host (Huff,
1969). Little host reaction has been observed in
association with the small endothelial schizonts of H.
f ringi 1 Iae . H. coIumbae , H. pa 1umbis , H â–  lophortyx, H.
ne 11 i on i s and ma c c a 1 1 um i (Khan and Fallís, 1969;
Moharrmed, 1965; Ahmed and Mohammed, 1977; 3aker, 1966a;
O'Roke, 1930; Sibley and Werner, 1984; Greiner, 1971).
Garnham (1966) reported cellular infiltration similar to
acute interstitial pneumonia in the alveolar septa of Rock
Doves with heavy, early infections of co1umbae, but did
not provide any details of the infection. Of the 3 genera
of avian haemosporidians, the exoerythrocy t i c stages of
Leucocytozoon have been documented as the most
pathogenic (Lund and Farr, 1965). However, most pathology

237
is associated with development of the large mega 1oschizonts
(Miller et a 1 ., 1983; Newberne, 1957).
During the earliest stages of infection with Hâ– 
me 1eagr i d i s , myopathy was restricted to isolated muscle
fibers. Sy 5 days post-infection, entire bundles of as
many as 5-10 adjacent myofibers were necrotic. The absence
of focal areas of inflammation and necrosis irrmediately
around f i rst-generat ion schizonts suggests that the
extensive necrosis was not due to the release of toxins by
developing parasites or to blockage or interference with
the local circulation. The extensive myopathy may have
been related to the size of the initial inoculum. The
total numbers of exoerythrocytic schizonts were
significantly fewer than in the pathogenicity experiment
where the total sporozoite dose was only 1/3 as large.
Perhaps many host cells were invaded by more than 1
sporozoite and were unable to support the development of
multiple schizonts. Their early death would explain the
low number of megaloschizonts that developed relative to
the infective dose. Wallace (1973) observed myositis
and myocarditis in mice that had been infected orally with
large numbers of Sarcocys t i s sporocysts. He found that the
myositis became apparent before sarcocysts developed and
was not associated with the early first-generation
schizonts. The myositis was also evident in experimentally
infected mice that failed to develop detectable sarcocysts,

238
suggesting that the early inflanmatory reaction may have
been to dead and dying host cells and parasites.
The host response surrounding 14-day-old and
17-day-old second-generation mega 1oschizonts is similar to
the host reactions to the large intramuscular cysts of
Sarcocystis. Leek et al. (1977) noted multifocal
perivascular inflamnatory infiltrates in the musculature of
lambs that were experimentally infected with Sarcocys tis.
The response was not associated with developing sarcocysts,
but was apparent around degenerating cysts. Mundy et al.
(1975) reported similar results. Other studies of
Sarcocystis and Hammondia have reported the necrosis and
mineralization of muscle fibers adjacent to developing
cysts (Cawthorn et al., 1984; Frenkel and Dubey, 1975).
Among haemosporidian parasites, the host response
to mega 1 oschizonts of me Ieagridis is similar to host
reactions to simondi . Miller et al. (1983), Desser
(1967), Cowan (1957) and Newberne (1957) reported the
presence of mixed inflammatory infiltrates composed of
mononuclear cells, heterophils, plasma cells and red blood
cells around both intact and ruptured mega 1 oschizonts as
well as necrotic changes in the surrounding host
tissue. Cowan (1957) and Miller et al. (1983) described
the spontaneous necrosis of Leucocytozoon
megaloschizonts in the absence of invading host cells. In
both studies, necrotic mega 1oschizonts were filled with an

239
amorphous, eosinophilic material that resembled the
material observed in necrotic mega 1 os ch i zonts of H.
me 1eagridis.
The mega 1oschizonts described by Nair and Forrester
(unpublished) in skeletal muscle of a naturally infected
Wild Turkey are identical in structure and size to
meg a I oschizonts of me 1eagrid i s from experimentally
infected turkeys. Many of the cysts observed by Nair
and Forrester contained disorganized masses of
basophilic material. Since the tissue had been frozen for
several weeks prior to fixation, degenerative changes in
the megaloschizont morphology may be an artifact of the
manner in which the tissue had been handled. The discovery
of these forms in a naturally infected Wild Turkey
indicates that the site of development and the
associated host response are not artifacts related to
the way the experimental turkeys were infected.
Mi 1tgen et al. (1931) noted the similarity between the
exoe rythrocytic stages of de s s e ri and Arthrocystis
ga 1 1 i , an organism of uncertain taxonomic status described
by Levine et al. (1970) in chickens from India, and
suggested that the 2 organisms may be synonymous. Other
descriptions of cyst-forming organisms have been
reported from a variety of avian hosts in several families
of birds. Most infections have been characterized by
the presence of large, intramuscular schizonts, similar in

240
morphology to the mega 1 osch¡zonts of Leucocytozoon
(Gardiner et al., 1984). In most cases, development of
megaIoschizonts occurred in the absence of circulating
gametocytes. Garnham (1973a, 1973b) suggested that
these were aberrant Leucocytozoon infections in abnormal
hosts. Gardiner et al. (1984) recently reviewed these
reports and described similar mega 1oschizonts in pen-reared
Northern Bobwhites from Ca1ifornia. The cysts described in
all the reports have a number of features in common,
including a hyaline wall of variable thickness, development
in muscle tissue and formation of numerous, spherical
merozoites approximately 1 um in diameter. Most have an
associated host myopathy.
The results of this study provide the first
experimental evidence that some of these organisms may
be haemoproteids . Gardiner et al. (1984) observed
pigmented gametocytes, morphologically similar to H.
1ophor t y x, in red blood cells of Northern Bobwhites that
died from an extensive myositis associated with developing
mega 1 oschizonts . However, since they observed small
schizonts with elongate zoites and hypertrophied host cell
nuclei in spleen tissue that were unlike any reported
schizonts of Leucocytozoon or Haemoproteus, they suggested
that the causitive organism may be a new member of the
Haemosporina. O'Roke (1930) conducted a detailed study of
H. 1ophortyx in California Valley Quail, but never reported

241
similar lesions. Instead, he found small schizonts in
endothelial cells of lung tissue. It is not clear from his
work, though, whether he examined histological sections of
skeletal, cardiac and gizzard muscle. The absence of
obvious gross lesions and cysts in the turkey with a
17-day-old infection of me 1eagr i d i s suggests that
megaloschizonts may be missed, particularly if their
development is not suspected. Since muscle tissue makes up
a large proportion of the total body mass, small numbers of
scattered cysts may be difficult to find and yet produce
enough gametocytes to infect a large proportion of the red
blood cells.
The elongate, f i rst-generat i on merozoites of Hâ– 
me 1 e a g r i d i s are similar to the merozoites contained in
small schizonts described by Gardiner et al. (1984). The
development of schizonts in the spleen of 1 turkey
during the pathogenicity experiment indicates that the
exoerythrocyt ic stages found by Gardiner et al. (1984) may
be part of the life cycle of lophortyx. Since
haemoproteids are believed to be specific to host family
(Bennett et al., 1982), Northern 3obwhites may be a
susceptible but aberrant host for lophortyx.
Clearly, additional experimental work with other species of
Haemop roteus is needed to determine the significance of
these exoerythrocytic stages in haemoproteid life cycles.

242
Pathology
Experimental infections of Haemoproteus meleagridis
produced a moderate to severe myositis and myocarditis
in domestic turkeys. The inflammatory reactions were
associated with the development of first and
second-generation schizonts. The lameness exhibited by
infected birds and the gross and microscopic lesions
were similar to naturally acquired infections of
Arthrocystis gall i (Levine et a 1 . , 1970; Opitz et al.,
1982). Both Levine et al. (1970) and Opitz et al. (1982)
noted extensive muscle necrosis, inflammation and
hemorrhage around the mega 1oschizonts. Opitz et al. ( 1982)
also observed dystrophic calcification of necrotic
muscle fibers and the formation of scar tissue. Neither
Opitz et al. ( 1 982 ) or Levine et al. (1970) reported
the thrombus formation or cicculatory disturbances that
were observed in this study. It is likely that they
developed in association with the acute and chronic
infla mm atory responses to intact and ruptured
mega 1 osch i zonts and the associated necrotic changes in
surrounding muscle fibers.
The distribution and morphology of second-generation
megalosch i zonts observed in the pathogenicity experiment
was identical to mega 1 oschizonts found during earlier
experimental infections. In addition, small,
thin-walled schizonts that contained both elongate

243
zoites and small, spherical pre-erythrocytic merozoites
were found in the spleen of I high dose bird that died
spontaneously. Other studies have reported the development
of exoerythrocytic schizonts of PI a s m o d i jum and
Leucocytozoon in reticular cells of the spleen (Huff, 1969;
Akiba et al., 1971). Host cell nuclei of the reticular
schizonts were not hypertrophied as is characteristic of
Leucocytozoon infections.
The reductions in growth and weight gain in
experimentally infected birds were dose dependent and most
pronounced between 1 and 3 weeks post-infection,
during development of second-generation
mega 1 oschizonts (Figures 37, 38). The onset of lameness
and anorexia in the high dose birds was approximately 1
week later than in the earlier series of experimental
infections and was probably associated with the
infla mm atory response to the second-generation
mega 1oschizonts. Decause high dose birds received fewer
than 1/3 as many sporozoites as birds that were infected to
study exoerythrocytic development, pathological changes
associated with development of the f i rst-gene r a t i on
schizonts may not have been as severe.
Following the crisis, all infected birds improved.
This was most evident among turkeys in the high dose group
that exhibited little significant weight gain between weeks
0 and 2, between weeks 1 and 3 and between weeks 3 and 4

244
post-infection (Figure 37). The surviving high dose birds
remained significantly smaller than control and low dose
turkeys throughout the course of the study (Figure 38).
Low dose birds were smaller than controls, but not
significantly so. If the sample size had been larger,
it is possible that significant effects on growth and
weight gain would have been detected for the low dose group
as well.
Few host effects were associated with the development
of the erythrocytic gametocytes, although these may have
been masked by the massive host response to the
mega 1oschizonts. Average hematocrit and hemoglobin values
were not significantly different for any of the 3
experimental groups at the crisis, or at the second peak in
parasitemia at 6 weeks post-infection (Figures 39, 41). A
significant drop in average hemoglobin concentration
occurred in the high dose group, 1 week after the crisis.
This drop corresponded to the rapid clearance of
parasitized red blood cells from the circulation and their
replacement with immature erythroblasts that had not
completely synthesized their total hemoglobin content
(Lucas and Jamroz, 1961). The absence of other significant
weekly differences in average hemoglobin concentration and
average hematocrit among the 3 experimental groups
indicates that removal of parasitized red cells was
balanced by the synthesis and release of erythroblasts.

245
When comparisons o£ average hematocrits were made by group,
the same trends were evident in each group, indicating that
significant differences within groups may be related to the
age of the birds (Figure 39).
Deposition of pigment in macrophages of the spleen,
liver and lungs began at approximately 22 DPI, when
maturing erythrocytic gametocytes began to develop
detectable pigment granules. The clearance of parasitized
erythrocytes from the circulation was probably accomplished
by phagocytic activity of these cells, as has been
described in infections of PIasmodi um (Taliaferro, 1941).
Most pigment was deposited in the spleen, where the
major elimination of the parasite population from the
peripheral circulation occurs (Taliaferro, 1941).
Follicular hyperplasia and enlargement of the spleen,
characteristic of other species of Haemopro t eu s and
P1 asmodium, also occurred (Becker et al, 1956; Russell
et al, 1943; Taliaferro, 1941).
The significant drop in plasma protein levels in
the low dose and big!) dose birds at 1 week
post-infection may have been related to the increase in
vascular permeability that accompanies acute
inflammatory responses (Figure 40) (Smith et al.,
1972). Mahrt and Payer (1975) did not find significant
changes in total serum protein during experimental
infections of calves with Saccocys tis fusiformis. However,

246
Leek et al. ( 1977) observed a significant drop in total
serum protein in lambs infected with Sa rcocys tis â–  The
decrease occurred during development of first-generation
schizonts in endothelial cells. Leek et al. suggested that
the drop in serum protein levels resulted from
glomerulonephritis associated with development of schizonts
in endothelial cells of the kidneys. Similar lesions were
not detected in infected birds examined in this experiment.
The increase in plasma protein concentrations in
the high dose group at 2 weeks pos t-i nf ect ion may have been
related to dehydration observed among these turkeys (Figure
40). Augustine (1982) noted similar increases in plasma
protein concentrations when turkeys were deprived of water
for periods of up to 72 hours. Wien birds were deprived of
both food and water, significant changes in plasma protein
concentrations did not occur. Since feed consumption
was not measured in the pathogenicity experiment, it is
unclear whether the high dose birds ceased eating and/or
drinking completely. The diarrhea associated with the
concurrent S a 1 inon ella infection may have acted in
conjunction with decreases in food and water consumption to
dehydrate the birds and concentrate total plasma proteins.
The increase in average plasma protein
concentration at 5 and 6 weeks post-infection may
reflect the synthesis of immunoglobulins (Figure 40).
Other studies of avian species of P1 a smodium and

247
Leucocytozoon have documented increases in parasite
specific immunoglobulins as the infections progressed
(Congdon et al., 1969; Mor ii, 1972).
3y o weeks pos t - i n f ec t i on , when the experiment was
ended, surviving birds had regenerating muscle fibers
and scarring associated with the destruction and removal of
mega 1 oschizonts (Figures 33 , 34 , 35 ). The persistence
of some meg a 1 oschizonts up to 8 weeks post-infection
suggests that they may be a source for relapses.
It is impossible to determine whether the high
(33%) mortality in the high dose group resulted from the H.
me 1eag rid i s infection alone, or whether the concurrent
SalmoneI 1 a infection was a significant factor. Studies of
the interactions between P^ be r g h ei and S a Imone11 a
typhimuriurn have shown that mice infected with both agents
died earlier than mice infected with either agent alone
(Kaye et al., 1965). Viens et al. (1974) had similar
results when they infected mice with P_^ y o e 1 i i and
Bo r de t e1 la pe r t u s s i . Cox (1978 ) suggested that the
synergism between Plasmodium infections and other
infectious agents may result from the imnunodepression that
often accompanies P1 a smodiurn infections. Similar
studies have not been conducted with Haemoproteus.
Because of the way they were housed, all birds in each
of the 3 experimental groups had an equal exposure to
Salmone11 a enteriditis. The close proximity of the battery

248
cage compartments to one another, the inevitable fecal
contamination that occurred in the food and «tet and
successful isolation of Salmone11 a from representatives of
each experimental group suggests that all the birds were
infected. Since birds vary in their output of
Sa 1 mone 1 la organisms from day to day, the low number of
Salmone11 a isolations from cloacal swabs of each group at 4
and at 3 weeks post-infection is not surprising (Williams,
1978). It is significant that the only birds to develop
clinical salmonellosis were those in the high dose group.
They exhibited signs of infection between 12 and 28 days
post-infection when stress from the lb me 1eag r i dis
infection reached its peak. Perhaps the Haemoproteus
infection weakened the high dose group sufficiently to make
t h em susceptible to Salmon ella and other secondary
bacterial and fungal infections as well. S a 1mo n e 1 la i s
frequently isolated from commercial feeds that use
animal products to boost protein levels (Williams, 1978).
It is likely that the turkeys used in the experiment
acquired their infections from the unmedicated game bird
chow they were fed.
Most authors have considered Haemoproteus to be a
relatively benign parasite (3ennett et al., 1982; Kemp,
1978; Fallis and Desser, 1977; Levine, 1961). Considering
how prevalent Haemoproteus is in many bird populations
(Bennett, 1982), the few isolated reports of pathogenic

249
effects in natural Haemoproteus infections have done little
to refute this view. The best documented pathology in
natural Haemoproteas infections was done by O'Roke
(1930) with his study of lophor t yx in California Valley
Quail. He attributed the morbidity and mortality he
observed to anemia caused by rupture of parasitized
erythrocytes and felt that the gametocytes made their host
cells more fragile and susceptible to rupture as they
passed through the capillary beds. The results of this
study indicate that me 1 e ag r i d i s can be severely
pathogenic in high doses and can have detectable effects on
growth and weight at low doses. The pathogenic effects
associated with the experimental infections was more
similar to host reactions to megaloschizonts of
L e ucocy tozoon than to the erythrocyte destruction and
anemia caused by PI a smodiurn.
Since the lesions in the naturally infected Wild
Turkey found by Nair and Forrester were similar to those of
the experimentally infected birds, 1C me 1eagridis may be a
cause of morbidity and mortality in Wild Turkeys. It
may only be significant in holoendemic areas such as
southern Florida and perhaps southern Texas where the
prevalence of the parasite ranges from 90-100% and rates of
transmission are high (Forrester et al., 1974; Cook et al.,
1966).

250
The low dose birds were infected with the sporozoites
contained in 5 infected eden i . A number of studies
of PIasmodiurn have shown that mosquitoes may inoculate only
a small percentage of their total number of salivary gland
sporozoites when they take a blood meal (Vanderberg,
1977). Similar studies have not been done with
ceratopogonids. However, since the prevalence of H.
me 1eagridis approaches 2% of the population of nulliparous
C. edeni at Fisheating Creek and since a bird may be bitten
by several thousand specimens of Cu I i coi des in a single
night, it is possible that intensities of natural infection
comparable to the low dose may be reached over a period of
several weeks. Additional research on the effects of
repeated, low level exposure of me 1eag ridis to
pen-reared Wild Turkeys would be helpful in determining the
possible significance of this parasite in the Wild
Turkey population.
Hos t Spec if i city
The present taxonomy of avian species of Haemoproteus
is based on the limited experimental evidence that suggests
that some species are specific to host family (Bennett
et al., 1982; Bennett et al., 1972). While this taxonomic
scheme provides a functional framework for dealing with the

251
large number o£ often poorly described species in this
genus, it is totally dependent on the maintenance of a
stable system of classification of the avian hosts.
Recently, the Committee on Classification and Nomenclature
of the American Ornithologist's Union (1983) published a
revised sixth edition of the Checklist of North American
Birds . For the first time since the publication of the
third edition in 1910, the coinmittee incorporated major
changes in the systematics of higher categories. These
changes were based on the conservative evaluation of
many new discoveries in the morphology, paleontology,
biochemistry, genetics and behavior of avian species
(Checklist of North American Birds, 1983). While the
status of many orders and families remained unchanged,
extensive revisions were made in others, e.g.
Passeriformes, Gall¡formes, that may require corresponding
revisions in haemoproteid taxonomy.
Prior to the 1983 revision of the Checklist of
North American Birds, the Order Gall¡formes was composed of
5 families: Cracidae (Curassows and Guans), Phasianidae
(Pheasants and Quail), Tetraonidae (Grouse), Me 1eagrididae
(Turkeys) and Numididae (Guineafowl). In the 1983 edition
of the Checklist of No r t h Ame rican Birds, only 2
families were recognized in the Order Gall¡formes: Cracidae
(Curassows and Guans) and Phasianidae (Pheasants, Quail,
Grouse, Turkeys, Guineafowl). The families Phasianidae,

252
Tetraonidae, Me 1eagr¡didae and Numididae were lowered to
subfamily status and the quail were separated from the
pheasants and elevated to form the subfamily
Odontophorinae.
Prior to the major taxonomic revisions in the Order
Galliformes, results of the host specificity experiment
would have invalidated assumptions about the host
specificity of avian haemoproteids and provided evidence
that the current taxonomy of the genus Haemoproteus may be
flawed. Instead, the finding that rL me 1eagridis can be
transmitted experimentally between Turkeys (subfamily
Me 1 eagridinae), Chuckars (subfamily Phasianinae, tribe
Perdicini) and Ring-necked Pheasants (subfamily
Phasianinae, tribe Phasianini), does not invalidate the
basic assumptions of current haemoproteid taxonomy.
Unfortunately, a true test of the taxonomic scheme was not
conducted since avian hosts outside of the current
Phasianid family were not included in the experiment.
However, the conflicting interpretations, dependent on host
classification, illustrate the dangers inherent in
basing the taxonomy of 1 unrelated group of organisms on
that of another.
The results of the host specificity experiment provide
some support for the current revisions within the Order
Galliformes. Nolan et al. (1975) found that rabbit
antibodies to 9 proteins purified from domestic

253
chickens reacted as well to their protein counterparts
from turkeys as they did to their protein counterparts from
Ring-Necked Pheasants. They briefly reviewed other
evidence from hybridization experiments, chromosome
studies, electrophoretic and inmunolog i ca I experiments and
anatomical studies that supported the similarities between
turkeys and other Phasianids. Since cell penetration
by some sporozoan parasites may be mediated by highly
specific cell surface receptors (Miller et al., 1978), the
parasites can, in a sense, be considered as highly specific
probes. The successful experimental transmission of H.
me 1e a g ridis from a turkey to a Ring-necked Pheasant and
a Chuckar suggests that all 3 species may have similar cell
surface receptors. Whi 1e these similarities may be of
minor taxonomic significance, they support the data showing
close biochemical similarities among these species.
The family Phasianidae presently contains members of 4
former avian families. Prior to the revision, Bennett
et al. (1982) reported 8 valid species of Haemoproteus
from the family Phasianidae - c h u c ka r i , rL chap ini,
H. ammo pe r d i s , f-K santosdiasi , bambú s i co 1 ae , HL
1ophor t yx, H. or a ta s i and rK ri 1e yi , 2 valid species of
Haemoproteus from the family Numididae - pratasi and
si 1 va i , 2 valid species from the family Tetraonidae - H.
mansoni (= rL canachites) and rK s t ab 1 e r i and 1 valid
species from the family Me 1eagr i didae - me Ie ag ridis.

254
Four species, fh chukari, H. coturnix, sal 1inarum and H.
perdix, were reported as nomen nuda. Bennett et al.
( 1 982 ) lists ba 1 four i from Guineafowl as valid, but
descriptions and figures of tilis species are absent from
the cited references. It will also be considered a
nomen nudum. Under the current system of haemoproteid
taxonomy, these species must be morphologica11y distinct
from one another to retain their status as separate taxa.
Haemoproteus me 1eagridis is most similar to rU man soni
and 1-L s t a b 1 e r i of Grouse (subfamily Tetraoninae).
Gametocytes of other species of Haemoproteus from the
subfamilies Phasianinae and Numidinae are halteridial
and only partially encircle the host cell nucleus. The
geographical range of ma n s o ni over laps mo s t of the
geographical range of me 1eagridis, while rh stableri has
been reported only from Ruffed Grouse in Montana (White and
Bennett, 1979). Greiner and Forrester (1980) discussed the
similarities between rh mansoni and rh meleagridis and
stated that the number of pigment granules in circumnuc1 ear
gametocytes of mansoni was significantly fewer than
in circumnuc1 ear gametocytes of me 1eagridis . They also
noted that the number of pigment granules in circumnuclear
forms of mansoni was fewer than in halteridial
forms. By contrast, the number of granules increased as
gametocytes of fh me 1eagridis became circumnuc1 ear.

255
Haemoproteus stab1er i can be distinguished from
me 1e a 3 cidis and man son i by the amoeboid margins of
mature gametocytes, their significantly fewer pigment
granules and the higher host cell nuclear displacement
ratio (White and Bennett, 1979; Greiner and Forrester,
1980).
The strong morphological similarities between
h a etnop r o t e i d s of Grouse and me 1 e a g r i d i s may be
significant. Bennett (1960) suggested that the host
behavior, habitat and corrmunity may be more important in
determining the host range of avian haemoprote i ds than
specificity of the parasites themselves. Since the
distribution of Ruffed Grouse coincides with that of the
Wild Turkey throughout the Appalachian Mountains of the
eastern U.S., transmission of fR mansoni and rR me 1eagridis
between either host may be possible. Fall is and Bennett
(I960) reported the sporogony of man soni in species
of Cu 1icoi des from Canada. It is likely that species of
C u 1 i c o i d e s are vectors of man son i throughout the
range of its host. Studies of the experimental cross
transmission of rK me 1 e agridis and man son i between
Ruffed Grouse and turkeys are needed to determine the
validity of both species of Haemoproteus â– 
The results of the morphometric analysis of H.
meleagridis in turkeys, the Ring-necked Pheasant and the
Chuckar demonstrate that few significant changes in

255
parasite morphology occurred. The discriminant analysis of
parasite and host cell variables was unable to correctly
classify scores from macrogametocytes, mi erógame tocytes and
host cells infected with macrogametocytes to their
respective host species. This indicates that the
garnetocytes of mel eagr i d i s and their associated changes
in host cells were essentially identical in each of the
3 host species. The discriminant analysis was more
successful in classifying host cells infected with
mi crogametocytes and correctly identified 100% of the
Ring-necked Pheasant scores. This was probably related to
the much greater lateral displacement of the host cell
nucleus that occurred in host cells of this species.
Morphometric studies of species of Leucocytozoon have
also failed to show significant morphological variation in
garnetocytes of a single species in different species of
hosts (3ennett and Campbell, 1975; Greiner and ivocan,
1977). These findings led Bennett and Campbell (1975)
to synonymize a number of mo rphologically similar
species reported from the same host family. These
synonymies were based on experimental evidence that most
species of Leucocy tozoon are specific to host family and
illustrate the importance of studies of
cross-transmission. Bennett and Campbell (1975) found that
garnetocytes of d ub r e u i 1 i , L. fringillinarum and the
round garnetocytes of si mo n di could not be separated

257
on the basis of measurements alone. All 3 species were
retained because experimental studies had demonstrated
their specificity to particular host families. In view of
the recent revisions in avian taxonomy, similar studies of
host specificity of avian haemoproteids, e . g . H â– 
me 1eag ridis and man soni, should be conducted before
synonymies are proposed.
Fine Structure
Mature Gametocytes
Mature gametocytes of me 1 eagridis were similar
in fine structure to gametocytes of other species of
Haemoproteus . Ultrastructural features of the
cytostoine, nucleus and nucleolus, mitochondria, osmiophilic
bodies, cytoplasmic ribosomes and endoplasmic reticulum
were essentially identical to gametocytes of H.
co 1 umba e , v e 1 an s and me t c h n i k o v i (Bradbury and
Roberts, 1970; Sterling, 1972; Sterling and Aikawa,
1973; Bradbury and Trager, 1968a; Desser, 1972a).
Differences from other species of Haemoproteus were
apparent in the pellicle of me 1eagridisâ–  Studies of
gametocytes of other species have shown that the
pellicle is composed of 3 layers - 2 outer unit

253
membranes and a thickened, ostniophilic inner membrane
composed of 2 unit membranes in close apposition to one
another (Bradbury and Roberts, 1970; Sterling, 1972;
Sterling and Aikawa, 1973). By contrast, the inner
membrane of mature gametocytes of me 1eagrid i s consisted
of a single unit membrane that was thicker and more
osmiophilic than the outer 2 membranes. It is unlikely that
the difference was a result of poor fixation, since
other membranous structures within the gametocytes and
their host cells were well preserved.
Sterling and Aikawa (1973) suggested that the
double inner membrane of gametocytes of co1umbae was
a remnant of the intramembranous complex of merozoites.
3ased on observations of PIasmod i urn spp., they suggested
that this 2-layered complex was retained in merozoites that
had recently invaded a red blood cell if the merozoite was
to become a gametocyte. De differentiation of the
complex occurred if asexual, schizogonic stages developed.
The similarities in structure and development among most
haemosporidian parasites make it unlikely that the
thickened, inner pellicle of me 1eagridis had a different
origin. Perhaps the 2-layered intramembranous complex
of merozoites of me 1 e a g t i d i s fused to form the
single, thickened layer that was observed in gametocytes.
Other studies of macrogametocytes of Haemoproteus and
Leucocytozoon have noted the presence of amorphous,

259
moderately electron dense material within the dilated
cisternae o£ the endoplasmic reticulum (Sterling and
Aikawa, 1973; Bradbury and Roberts, 1970). Desser et
al. (1970b) suggested that it was the precursor to the
crystalloid material observed in ookinetes, oocysts and
sporozoites of si mo n di â–  The continuity between the
endoplasmic reticulum and the inner, osmiophilic layer
of the pellicle of macrogametocytes of me 1eagridis,
indicates that this material may also play some role in the
changes in pellicular structure that occur prior to
their release from the host cell. Other workers have
suggested that the osmiophilic bodies may have a similar
function and aid in dissolving the host cell membrane
during gametogenesis (Aikawa et al., 1969; Rudzinska and
Trager, 1968).
Gametogenesis
Early changes in the fine structure of gametocytes of
H■ me Iea»ridis during the initial stages of gametocyte
maturation were identical to changes observed in
maturing gametocytes of co 1 umbae , metchn i kov i and
ve 1ans (Bradbury and Trager, 1968a; 3radbury and Trager,
1968b; Sterling, 1972; Aikawa and Sterling, 1974a; Desser,
1972a). In all species studied to date, the gametocytes
round-up within their host cells. Soon afterward, the
outer layer of the pellicle detaches from the outer surface

260
of the gametocyte in sheets and whorls and floats free
in the host cell cytoplasm. Sterling (1972) did not
observe the detachment of the outer layer from gametocytes
of metc'nnikovi , but did note that it was missing once
the gametocytes became extracellular. Desser (1972a)
described mu 1ti I aminar, membranous structures external
to maturing, intracellular gametocytes of ve 1ans, but
failed to associate their appearance with the loss of
the outer layer of the pellicle in extracellular
gametocytes. He suggested instead that they were
altered bands of microtubules from host red blood cells
that normally aid in maintaining the shape of the cell.
Micrographs published by Bradbury and Roberts (1970) and
Aikawa and Sterling (1974), as well as observations from
this study, leave little doubt that these membranous sheets
and whorls originated from the outer layer of the
gametocyte pellicle. This process appears to be limited to
the genus Haemoproteus and has not been reported from
studies of PI a smodium o t Leucocytozoon (Aikawa and
Sterling, 1974a). Ribosome-1ike granules similar to those
observed on some membranous whorls in this study were also
observed by 3radbury and Roberts (1970) on whorls that
detached from gametocytes of co1umbae. The significance
of the granules and their origin are unknown.

261
Mac rosametogen es is . Few detailed ultrastructural
observations of the maturation of ma erógame tes have been
made for any species of Haemoproteus â–  3radbury and Trager
(1968a) provided the most detailed description. During the
initial steps of gametocyte maturation in columbae, they
noted the detachment of the endoplasmic reticulum from the
nuclear envelop and the elongation of the
macrogametocyte nucleus from a spherical to an ellipsoidal
shape. The changes in nuclear shape were accompanied by
the appearance of an intranuclear spindle that converged on
an electron dense plaque in the nuclear membrane. Two
"atypical" centrioles composed of a circle of 9, single
microtubules rather than the conventional circle of 9,
triple microtubules were embedded in an electron dense
material. They appeared in the cytoplasm next to the
electron dense plaque. Approximately 10 minutes after the
start of gametocyte maturation, the nucleus constricted
twice to form 2, detached maturation bodies. The
intranuclear spindle as well as the nucleolus were retained
by the nucleus.
A similar process appeared to occur in maturing
inacrogametocytes of H_j_ meleagridis , although
observations are limited. Unfortunately, gametocyte
maturation was not followed longer than 3 minutes, so
the formation of maturation bodies was not observed.
Atypical centrioles were not found in association with the

262
electron dense thickening and associated intranuclear
spindle in 1 macrogaraetocyte . Since serial sections
were not cut, it is likely that they may 'nave been missed.
A single cytostoine normally occurs in mature
gametocytes of columbae and rL metchnikovi (Bradbury and
Roberts, 1970; Sterling, 1972). Gallucci (1974a) and
Bradbury and Trager (1968a) reported "internal cytostomes"
and a single, non-functional peripheral cytostome within
mature and immature macrogametes of H^ columbae. The
internal cytostomes consisted of 2 electron dense rings,
but were located in the interior of the cell rather than in
the external pellicle. Other observations of cytostomes in
maturing macrogametes and exflagel bating
mi erógametocytes have not been reported. The presence
of cytostomes in developing gametes of me 1eagrid i s
indicates that they persist as part of the pellicle
throughout development of the gametes. However, their
small size and the absence of associated food vacuoles
indicates that they are non-funct iona 1 .
Microgametogenes i s . A number of studies of the
microgametogenesis of avian haemoproteids have been
conducted. Bradbury and Trager (1968a, 1968b), Sterling
(1972) and Aikawa and Sterling (1974a) studied co1umbae
and metchnikovi. They reported the development of
axon ernes composed of 9 peripheral microtubules that

263
surrounded 2 central microtubules during the earliest
stages of exf1 age I 1 at i on . These were located free in
the cytoplasm in various stages of assembly. One end of
developing axonemes was usually associated with a dense
plaque in the nuclear membrane that usually had associated
intranuclear tubules. Sterling (1972) reported an atypical
centriole embedded in electron dense material adjacent
to the plaque. He observed the attachment of axonemes
to basal bodies that v/ere associated with the atypical
centriole. Aikawa and Sterling (1974a) found that the
intranuclear microtubules extended across the nucleus of
ex flagellating gametocytes of co1umbae to another
electron dense plaque on the opposite side of the
organelle. They observed the condensation of electron
dense material around the base of the plaques and suggested
that this material was eventually incorporated into the
microgamete nucleus. As axonemes began to bud from the
exterior of m i c r ogaine t oc y t e s , Aikawa and Sterling
(1974a) described the protrusion of a portion of the
mi c r ogame t ocy t e nucleus to the base of the bud. They
presented micrographs indicating that a portion of the
nucleus, still bound by a nuclear membrane, detached
from the mi erógametocyte nucleus and became incorporated
into the exf1 age 11 ating microgamete as a spiral around the
axonerne. Sterling (1972) described a similar process
during the ex f 1 age 1 1 a t i on of me t chn i Itov i .

264
Bradbury and Trager (1968b) described the polarization
of in i c r oga me t ocy t e s of H^ col umbae into 2 halves - 1
that contained organelles such as axonemes, mitochondria
and food vacuoles and another that contained remnants of
the microgametocyte nucleus. They observed the
disintegration of the nuclear membrane and a
condensation of nuclear material around the bases of
developing axonemes. The condensed masses of nuclear
material were subsequently surrounded by a membrane and
incorporated into microgametes that "peeled" from the
mi erógametocy te.
Desser (1972a) described the formation of atypical
cent r¡oles and electron dense nuclear plaques in
mi c r ogame t ocy t e s of ve 1 an s . He did not observe the
polarization of exflagellating micro gametocytes
described by Bradbury and Trager (1968b), but did describe
the breakdown of the nuclear membrane and the subsequent
condensation of masses of chromatin around the bases of
developing axonemes.
Observations from this study are very limited, but
they are most similar to the process described by Aikawa
and Sterling (1974a). The nuclear membrane remained intact
in m i c r og ame t oc y t e s of me 1 eag r i d i s throughout the
process of exflagellation, but was often difficult to
discern. During the final budding of microgametes, the
nucleus appeared to be stretched to their base. The

265
polarization of mi crogametocy tes of me 1eagridis into
2 halves was not observed. Atypical centrioles were not
detected, but they may have been missed since serial
sections were not cut.
Microgametes of me 1eag rid i s were similar to
microgametes of other species of ilaemop roteus .
Microgametes of most other species contain a single axoneme
and a small, membrane-bound nucleus. Mitochondria have not
been observed (Aikawa and Sterling, 1974a; Sterling, 1972;
Desser, 1972a). Bradbury and Trager (1968a) observed 2
axonemes in mi c r ogame t e s of co 1 umbae , but these
observations have not been confirmed (Aikawa and Sterling,
1974a). The periodic striations observed in the central
microtubules of axonemes of me 1eagridis have been
reported in axonemes of me t c h nik o v i and si mo n di
(Sterling, 1972; Aikawa et al., 1970).
Oocyst s
Most ultrastructuraI studies of the oocysts of
haemosporidian parasites have been limited to avian and
mammalian species of PI a smodium (Mehlhorn et al., 1980;
Sinden and Strong, 1978; Canning and Sinden, 1973; Howells
and Davies, 1971; Terzakis, 1971; Terzakis et al., 1966;
Duncan et al., 1960) and to 3 species of Leucocy tozoon
(Desser and Allison, 1979; Wong and Desser, 1976;
Desser, 1972c). In spite of the importance of comparative

studies in elucidating the taxonomic and phylogenetic
relationships among the Haemosporina, only 1 study oE
the oocysts of Haemoproteus has been published (Sterling
and DeGiusti, 1974). These workers found that H.
me t c h nik o v i , a chelonian parasite that develops in deer
flies, shares features of its sporogonic development
w i t h P 1 a smod i urn and Leucocy t o zoo n that neither of
these2o genera hold in comuon.
Differentiation of the oocyst. The sporogony of avian
and mammalian species of P1 a smod i urn follows a number of
similar steps (Mehlhorn et al., 1980; Sinden and Strong,
1978; Garnhan et al., 1969; Terzakis et al., 1967; Terzakis
et al., 1966; Duncan et al., 1960). After penetrating the
midgut, the ookinete rounds-up under the basement
membrane. Mehlhorn et al. (1980) found the remains of
polar rings and mictonemes derived from the apical complex
of the ookinete at the periphery of early FG ga 1 1 inaceum
oocysts. Similarly, Garnham et al. (1969) found
remnants of this complex in a dimple or small groove In the
peripheral cytoplasm of early oocysts of P^ b e r g h ei
yoe1 ii. Studies of older PI asmodium oocysts have failed to
find evidence that these organelles persist more than a few
days. Among other Haemosporina, the remnants of apical
organelles from the ookinete have been observed in
early, but not later oocysts of t awaki (Des ser and

267
Allison, 1979). Sterling and DeGiusti (1974) observed
these organelles throughout the development of oocysts
of metchnikovi and in the residual body after sporozoite
differentiation was complete. Apical organelles derived
from the ookinete of meleagridis appear to persist in
the cytoplasm throughout the life of the oocyst. There is
no evidence that any of them are reused in the formation of
sporozoites (Garnham et a 1., 1969).
The early stages of sporozoite formation in PI a stood i um
are initiated by vacuolization of the peripheral oocyst
cytoplasm. This process also occurs in early oocysts of H.
metchn i icov i , but has not been described in any of the 3
species of Leucocytozoon that have been studied. Terzakis
et al. (1966) suggested that the vacuoles inay develop after
a local change in ionic concentration in the peripheral
cytoplasm causes water to cross the oocyst wall or,
alternatively, after the oocyst cytoplasm secretes fluid at
the oocyst periphery. In support of the latter, Sinden and
Strong (1973) described the formation of vacuoles in
oocysts of falciparum by the fusion of membranous
vesicles that originated in the oocyst cytoplasm.
Electron dense, linear thickenings appear under the
membrane soon after vacuolization begins. The vacuoles
eventually coalesce to form cytoplasmic clefts which
subdivide the oocyst cytoplasm. Sinden and Strong
(1978) found that clefts in the oocysts of falciparum

268
originated from endoplasmic reticulum scattered in the
cytoplasm and from peripheral vacuolization of the oocyst.
Clefts originating from the endoplasmic reticulum were
covered on their cytoplasmic face by additional 2-layered
membranous sacs. These were lined on their innermost face
by a thin amorphous electron-dense coating analogous to the
electron, linear thickenings described under the plasma
membrane of other species.
As subdivision of the cytoplasm continues, budding
sporozoites develop under the electron dense thickenings.
Apical organelles including polar rings, micronemes and
subpe 11 icu1 ar microtubules differentiate as the sporozoites
bud from the sporoblast body. In contrast to
P 1 a s modiurn, metchnikovi does not undergo cleft
formation. Instead, budding sporozoites develop around the
periphery of the central sporoblast body (Sterling and
DeGius ti, 1974).
The differentiation of early oocysts of Leucocytozoon
occurs without vacuolization and cleft formation. As
sporozoites bud from electron dense thickenings under
the peripheral plasma membrane, the sporoblast body slowly
contracts (Desser and Wright, 1968; Desser, 1972c; Wong and
Desser, 1976; Desser and Allison, 1979). Sporozoite
differentiation and budding resembles P1 a smodiurn.
Leucocytozoon oocysts are smaller and develop fewer
sporozoites than PI asmodium oocysts. The cleft formation,

269
subdivision of the oocyst cytoplasm into many sporoblast
bodies and the endogenous or internal sporozoite budding
that occurs in P1 a smod i urn is probably an adaptation to
increase the surface area available for sporozoite
formation (Sinden and Strong, 1978).
Limited observations of me 1 eagr i d i s from this study
suggest that vacuolization and cleft formation do not
occur. Thus, this parasite closely resembles Leucocytozoon
in size of oocysts and number of sporozoites produced as
well as in contraction of the sporoblast body and
differentiation of the sporozoites.
Nuclear divisions. The nuclear events following
fertilization of haemosporidian parasites are still poorly
understood. Bano (1959) and Canning and Anwar (1963) found
evidence of post-zygotic meiosis in the early oocysts of
P1 a smodiurn. They observed what they believed to be 4
chromosomes in diploid oocysts and 2 in haploid oocysts in
stained preparations. Later studies with electron
microscopy have shown that the chromosomes do not
condense. The dark-sta in i ng nuclear masses observed
earlier may have been nucleoli (Mehlhorn et al., 1980)
or fragments of an enlarged digitate nucleus (Canning
and Sinden, 1973). Mehlhorn et al. (1980) observed nuclear
spindles in zygotes and in ookinetes of P^ ga 1 1 i naceum and

270
suggested that zygotic rather than post-zygotic ineiosis had
occur red.
Attempts to trace division of the nucleus during
differentiation of PIasmodiu n oocysts have been only partly
successful. This is largely because of the difficulty
in obtaining and interpreting serial sections through
the complex, multidigitate form of the nucleus. A
number of investigators have found that nuclear division
proceeds without the disappearance of the nuclear
membrane. Electron dense masses termed centrioiar plaques
(Howells and Davies, 1971), kinetic centers (Schrevel et
al., 1977) and spindle pole bodies (Kubai, 1975) appear in
cup-like invaginations of the nuclear envelope. These are
duplicated throughout the nuclear envelope and direct a
series of multiple, asynchronous mitotic divisions.
Each kinetic center is linked with another by a thin
band of dense material. Radiating from each are spindle
microtubules of different lengths. The shorter ones appear
to be attached to electron dense kinetochores. These
are believed to be attached to individual chromosomes
(Schrevel et al., 1977). Schrevel et al. (1977) counted 8
kinetochores at each halfspindle pole, suggesting that the
haploid chromosome number in P^ her ;he i may be as few as
4. Canning and Sinden (1973) estimated that the haploid
number could be as few as 5 and possibly as high as 10 from
kinetochore counts on their micrographs of P^ berghei.

271
Since most studies o£ oocysts of liaemospor i d i an
parasites are not based on serial sections, it is not clear
when final fragmentation of the large polyploid nucleus
occurs. Howells and Davies (1971) suggested that the
large, lobulate nucleus of be rghei may break up soon
after cleft formation subdivides the cytoplasm into
sporoblast bodies. Each nucleus may then undergo a
final division as opposing ends migrate into sporozoite
buds. Schrevel et a 1 . (1977) felt that budding and
fragmentation of the nucleus occurred at the same time
sporozoites budded from the sporoblast body. They also
suggested that the multiple mitoses that occurred prior to
sporozoite differentiation allowed numerous genetic
units to be positioned in the nuclear cortex for the
direction of sporozoite differentiation.
Studies of Leucocytozoon have indicated that
nuclear division may occur in a manner similar to
PIasmodiurn. Multiple kinetic centers have been observed in
the nuclear envelope of s imond i , L_^ dubreu i 1 i and L.
t a w a ki, but the sequence of events leading to final
fragmentation of the nucleus has not been studied in detail
(Desser, 1972c; Wong and Desser, 1976; Desser and Allison,
1979).
Kinetic centers have not been described in H â– 
metchnikovi, nor were they found in this study.
However, it seems likely that haemoproteids may undergo

272
nuclear divisions in a manner similar to ?1 a smod i urn and
Leucocytozoon.
Crystal 1 oidâ–  Crystalloid aggregations of electron
dense particles consisting of a lipo-protein complex
have been reported from the oocysts of Leucocytozoon, H.
me t c h n i!: o v i and from the early oocysts of EC ga 1 1 i naceuin
and P^ b e r g h e i (Trefiak and Desser, 1973). Trefiak and
Desser (1973) found that they they originate from amorphous
aggregations of electron dense material in the macrogametes
o f L_^ s i mond i . They suggested that the crystalloid
functions as an energy source since it appears to be
utilized quickly in rapidly growing oocysts of
P1 a smodium and is incorporated into the sporozoites of
Leucocy tozoon and me t chnikovi. The former have been
found to persist in their avian hosts for up to 11 days and
may require a reserve energy source to maintain their
viability (Khan et al., 1969).
Membrane-bound lipid inclusions have been reported in
the oocysts of !C cynomo 1 g i , EC ga 1 1 i naceum, EC f a 1 c i par um,
L . s i mo n d i , dub r eu i 1 i and 1^ tawak i (Terzakis, 1971;
Terzakis et al., 1966; Sinden and Strong, 1978; Desser,
1972c; Wong and Desser, 1976; Desser and Allison,
1979). They were not reported in oocysts of !C metchnikovi
(Sterling and DeGiusti, 1974). Generally, lipid inclusions

273
tend to be smaller in oocysts of those species that form
prominent crystalloids.
The absence of a crystalloid in 3-day-old and
6-day-old oocysts of me 1e a g ridis suggests that the large
lipid-like inclusions observed in this study may
function as energy reserves. Observations of earlier
oocysts are needed to determine whether a crystalloid is
present in this parasite.
Mega 1oschizon t s
The thick laminated cyst wall and the process of
inerozoi te formation within mature mega 1 osch i zonts of H,
me 1eagr i d i s appears to be unique among haemosporidian
parasites. Observations of meg a 1 oschizonts of L.
s i mo nd i and the meg a 1 o s c h i zo n t s of the unidentified
haemosporidian from Northern Bobwhites are most similar to
those of this study. Desser (1970a) reported a capsule
around mega losch i zon t s of s i mo n d i composed of a meshwork
of reticular fibers and an outer fibrous layer. The
capsule was external to the host cell plasma membrane.
Desser and Fallís (1967) suggested that it was primarily of
host origin and probably secreted by fibroblasts that were
often found surrounding the mega Ioschizonts and their host
cells. Gardiner et a 1 . (1984) found that
mega 1oschizonts of the parasite of Northern Bobwhites were
surrounded by a moderately dense, amorphous wall that

274
had villous processes extending £rom their external
surfaces. They also noted that the megaloschizonts were
divided into thick-walled compartments resembling tubes.
Megaloschizonts of me 1 e ap,r i d i s differ from those
observed by Desser (1970a) and Gardiner et al. (1984). The
megaloschizonts observed in this study were
extracellular and had thick, laminated walls composed of
electron-dense, granular material. Material examined by
Gardiner et al. (1984) was fixed initially in buffered
formalin and then post-fixed with osrni um-d i chromate and
processed for electron microscopy. The poor
preservation of cellular detail in their samples make
comparisons with the megaloschizonts of me 1e agridis
difficult. However, the villous processes on the cyst
walls and compartmentalization of megaloschizonts by the
thick wall were never observed in megaloschizonts of H.
me 1eagridis .
Desser (1970a) described merozoite formation in
megaloschizonts of simo n di by fragmentation of the
cytomere cytoplasm. By contrast, Gardiner et al. (1984)
reported that merozoite development in megaloschizonts from
Northern Bobwhites occurred by budding into an interior
vacuole within the cytomere cytoplasm rather than by the
protrusion of developing merozoites from the cytomere
surface. Neither process was observed in
megaloschizonts of me 1eagridis.

275
Mature merozoi tes of me 1 e a 3 rid i s are very
similar to merozoites of other ha emospor i d i an
parasites. All have a specialized anterior end containing
3 polar rings, a pair of electron dense rhoptries and small
micronemes (Aikawa and Sterling, 1974b). Cytostomes
have been reported in merozoites of P1 a smodiurn and
Haernop roteus , but have not been observed in merozoites
of species of Leucocytozoon (Aikawa and Sterling, 1974b).
They were not observed in merozoites of me 1eagridis, but
observations are limited. Other organelles, including the
nucleus, mitochondria and pellicle, were similar to
merozoites of other haemosporidian parasites (Aikawa and
Sterling, 1974b).
The large, e1ectron-1ucent vacuoles observed in mature
merozoites of me 1eagridis have not been reported from
other haemosporidian merozoites (Figures 90, 91). The
vacuole persists during development of the early
gametocytes and is clearly visible in Giemsa-stained blood
films. It disappears by the time gametocytes reach
maturity. The function and origin of this organelle are
unknown.

LITERATURE CITED
Acton, H.'H. , and R. Knowles. 1914. Studies on the
halteridium parasite of the pigeon, Haemoptoteus
columbae Ce 11i and San Felice. Indian J. Med. Res. TT
663-690.
Ad i e , H.A. 1915. The sporogony of Haeinoproteus co 1 umbae.
Indian J. Med. Res. 2: 671-680.
Ad i e , H.A. 1924. The sporogony of Haeinoproteus co 1 umbae.
3ull. Soc. Pathol. Exot. 17: 605-613.
Adie, H.A. 1925. Nouvelle recherches sur la sporogonie de
Haemoproteus columbae. Arch. Inst. Pasteur
Algér. 3: 9-1'S.
Ahmed, F.E., and A.H.H. Mohammed. 1977. Schizogony in
Maemooroteus columbae Kruse. J. Protozool. 24:
389-393.
Aik a w a , M. , C.G. Huff and C.P.A. Strome. 1970.
Morphological study of microgametogenesis of
Leucocytozoon simondi. J. Ultrastruct. Res. 32: 43-68 .
Aikawa, M., and C.R. Sterling. 1974a. High voltage
electron microscopy on rn ¡erogametogenes is of
Haemoproteus columbae. Z. Zellforsch. 147: 353-360.
Aikawa, M., and C.R. Sterling. 1974b. Intracellular
parasitic protozoa. Academic Press, New York. 76 pp.
Aikawa, M. , C.G. Huff and H. Sprinz. 1969. Comparative
fine structure of the gametocytes of avian, reptilian
and mammalian malarial parasites. J. Ultrastruct.
Res. 26: 316-331.
Akey, B.L. 1981. Mortality in Florida wild turkey poults
(M e1e a g ri s gal lopavo o s c e o 1 a). M.S. Thesis.
University of Florida, Gainesville, Florida. 78 pp.
276

277
Ak iba, ¡i., S. Inui and R. Ishitani. 1971. Morphology and
distribution of intracellular schizonts in chickens
experimentally infected with Akib a caulleryi.
Nat. Inst. An. Hlth. Quart. 11: 109—121.
American Ornithologists' Union. 1983. Checklist of
North American Birds. 6th edition. American
Ornithologists' Union, Washington, D.C. 877 pp.
Applegate, J.E. 1970. Population changes In latent malaria
infections associated with season and
corticosterone treatment. J. Parasitol. 56: 439-443.
Aragao, H.B. 1908. Sobre o cyclo evolutivo e a transmissao
do Haemoproteus columbae. Rev. Med. (Sao Paulo). 11:
416-419.
Aragao, H.B. 1916. Pesquizas sobre o "Haemoproteus
columbae"â–  Bras. Med. 30: 353-354, 361-362.
Atkinson, C.T., E.C. Greiner and D.J. Forrester.
1983. Experimental vectors of Ha emoproteus
meleagridis from wild turkeys in Flo r i da . .1. WTl Dis. Í9: 366-368.
Augustine, P.C. 1982. Effect of feed and water deprivation
on organ and blood characteristics of young turkeys.
Poultry Sci . 61 : 796-799.
Ayala, S.C., J.M. Ramakka, V.F. Ramakka and C.E. Varela.
1977 . Haemoproteus , Plasmodium and hippoboscid
ectoparas i tes of Co 1 umb i~añ wi 1 d doves. Rev. Inst.
Med. Trop. Sao Paulo. 19: 411-416.
Baker, J.R. 1957. A new vector of Ha emop roteus
columbae in England. J. Protozool. 4: 204-208.
Baker, J.R. 1963. Transmission of Haemoproteus sp. of
English wood-pigeons by Ornithomyia avicularla. J.
Protozool. 10:461-465.
Baker, J.R. 1966a. Haemoproteus pa 1umbis sp.nov. (Sporozoa,
Haemosporina ) of the Eng 1ish Wood-Pigeon Co Iumb a p.
pa Iumbus. J. Protozool. 13: 515-519.
Baker, J.R. 1966b. The host-restriction of Haemoproteus
sp. indet. of the wood-pigeon Co I u mb a pT
pa 1umbusâ–  J. Protozool. 13: 406-408.

273
Saker, J.R. 1963. The host restriction o£ Haemoproteus
columbae. J. Protozool. 15: 334-335.
Baño, L. 1959. A cytological study o£ the early oocysts of
seven species of PI a smod i um and the occurrence o£
post-zygotic meiosisC Par asito 1ogy. 49: 559-535.
Barnard, D.R. and R.H. Jones. 1980. Die! and seasonal
patterns of flight activity of ceratopogonids in
northeastern Colorado. Environ. Entomol. 9: 446-451.
Bates, M. 1949. The natural history of mosquitoes. The
Macmillan Co. New York. 379 pp.
Becker, E.R., W.F. Hollander and W.H. Patti 1 1o. 1956.
Naturally occurring P1 a smodium and Haemop r oteus
infection in the common pigeon. J. Pa r a sito 1. T2 :
474-478.
Bennett, G.F. 1960. On some ornithophilic
blood-sucking Díptera in Algonquin Park, Ontario,
Canada. Can. J. Zool. 38: 377-389.
Bennett, G.F., and A.G. Campbell. 1972. Avian
Haemoproteidae. I. Description of Haemoproteus fa 11 isi
n.sp. and a review of the haemoprote i as ol the tamily
Turdidae. Can. J. Zool. 50: 1269-1275.
Bennett, G.F., and A.G. Campbell. 1975. Avian
Leucocytozo idae. 1. Morphometric variation in three
species of L eucocytozoon and some taxonomic
implications. Can. Zool. 53: 800-312.
Bennett, G.F., and R.F. Coombs. 1975. Ornithophilic
vectors or avian hematozoa in insular
Newfoundland. Can. J. Zool. 53: 1241-1246.
Bennett, G.F., and A.M. Fallís. 1960. Blood parasites of
birds of Algonquin Park, Canada, and a discussion of
their transmission. Can. J. Zool. 38: 261-273.
Bennett, G.F., P.C.C. Garnham and A.M. Fallís. 1965. On
the status of the genera Leucocy tozoon Ziemann 1898
and Ha emoproteus Kruse, 18 9 0 (Ha emos po rid i d a :
Leucocytozoidae and Haemoprote i dae) . Can. J. Zool.
43: 927-932.

279
Bennett, G.F., N.O. Okia, R.G. Ashford and A.G. Campbell.
1972. Avian haemoproteidae. II. Haemoproteus
enucleator sp.n. from the kingfisher Ispidina picta
(3oddaer t). J. Parasitol. 58: 1143-1147.
Bennett, G.F., M. Whiteway and C. Woodworth-Lynas . 1982.
A host-parasite catalogue of the avian haematozoa.
Mem. U. Newfoundland Occ. Pap. Biology. 5. Dept. of
Biology, Memorial University of Newfoundland, St.
John's, Newfoundland.
Bid1 ingmeyer, W.L. 1961. Field activity of adult
Cu 1icoi des furens . Ann. Ent. Soc. Amer. 54: 149-156.
B i d 1 ingmeyer , W.L. 1969 . The use of logarithms in
analyzing trap collections. Mosq. News. 29: 635-640.
Bierer, B.W., C.L. Vickers and J.B. Thomas. 1959. A
parasitism in turkeys due to a Haemoproteus-1 ike blood
parasite. J. Am. Vet. Med. Assoc"! 1 35 : 1 81-1 82.
Blanton, F.S., and W.W. Wirth. 1979. The sandflies
( C u 1 i c o i d e s ) of Florida (Díptera:
Ceratopogonidae) . Florida Dept, of Agriculture and
Consumer Services. Gainesville, Florida. 204 pp.
Bowman, J.A., C.E. Hill and R.Q. Burleson. 1979. Seasonal
movements of restocked wild turkeys in North
Carolina. Proc. Ann. Conf. S.E. Assoc. Fish &
Wildl. Agencies.' 33: 212-223.
Bradbury, P.C., and B.B. Gallucci. 1971. The fine structure
of differentiating merozoites of Haemoproteus columbae
Kruse. J. Protozool. 18: 679-683"!
Bradbury, P.C., and B.B. Gallucci. 1972. Observations
on the fine structure of the schizonts of
Haemoproteus co1umbae. J. Protozool. 17: 405-414.
Bradbury, P.C., and J.F. Roberts. 1970. Early stages
in the differentiation of gametocytes of Haemoproteus
co 1 umbae Kruse. J. Protozool. 17: 405-413"!
Bradbury, P.C., and W. Trager. 1968a. The f ine structure
of the mature gametes of Haemoproteus coIumbae Kruse.
J. Protozool. 15: 89-102.
Bradbury, P.C., and W. Trager. 1968b. The fine
structure of m i crogametogenes i s in Haemop roteus
columbae Kruse. J. Protozool. 15: 700-712.

280
Bray, R.S. I960. Observations on the cytology and
morphology of the mammalian malaria parasites. 1. A
process of apparent plasrnotomy in the preerythrocyt ic
phase of Laverania falciparum. Riv. Parassit. 21:
267-276.
Bray, R.S., and P.C.C. Garnham. 1962. The
Giemsa-colophoniurn method for staining protozoa in
tissue sections. Ind. J. Malar. 16: 153-155.
Canning, E.U., and M. Anwar. 1968. Studies on meiotic
division in coccidial and malarial parasites. J.
Protozool. 15: 290-298.
Canning, E.U., and R.E. Sinden. 1973. The organization
of the ookinete and observations on nuclear
division in oocysts of Plasmodium b e r g h e i .
Parasitology. 67: 29-40.
Cawthorn, R.J., A.A. Gajadhar and R.J. Brooks. 1 984.
Description of Sarcocystis rauschorum sp.n. (Protozoa:
Sarcocystid a e ) with experimental cyclic
transmission between varying lemmings (Pierostonvx
richardsoni) and snowy owls (Nyctea scandiaca).
Ca~:i; zobi . -
62:
s nowy
217-225.
Cawthorn, R.J., G.A. Wobeser and A.A. Gajadhar. 1983.
Description of Sarcocystis campes tris sp.n. (Protozoa:
Sarcocystidae): a parasite ol the badjer Tax idea taxus
with experimental transmission to the Richardson's
ground squirrel, Spermophilus ri chardson i i . Can.
J. Zool. 61: 370-377.
Chernin, E. 1952. The epi zooti o 1ogy of Leucocy tozoon
si mo n di infections in domestic ducks Tn northern
Michigan. Am. J. Hyg. 56: 39-57.
Climatological Data: Florida. 1982. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.
Climatological Data: Florida. 1983. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.

281
Climatological Data: Florida. 1984. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.
Coatney, G.R. 1933. Relapse and associate phenomena in the
Haemoproteus infection of the pigeon. Am. J. Hyg. 18:
133-1 ¡50.
Congdon, L.L., J.N. Farmer, B.M. Longenecker and R.P.
Bre i tcnbach. 1969. Natural and acquired
antibodies to Plasmodium 1 ophu rae in intact and
bursaless chickens. II. I mmunoflore scent studies.
J. Parasitol. 55: 817-824.
Cook, R.S., D.O. Trainer and W.C. Glazener. 1966.
Haemoproteus in wild turkeys from the coastal bend of
south Texas. J. Protozool. 13: 588-590.
Cowan, A.B. 1957. Reactions against mega 1osch i zonts of
Leucocytozoon simondi Mathis and Leger in ducks. J.
Infect. Dis. 100: 82-87.
Cox, F.E.G. 1978. Concomitant infections, in "Rodent
Malaria". R. Ki11 ick-Kendrick and W. Peters, eds.
Academic Press, New York. pp. 309-343.
Danilewsky, B. 1 889 . La parasito 1og i e comparie du sang.
I. Nouvelles recherches sur les parasites du sang des
oiseaux. Kharkov. 93 pp.
Desser, S.S. 1967. Schizogony and gametogony of
Leucocytozoon simondi and associated reactions in the
avian nost. J. Protozool. 14: 244-254.
Desser , S.S. 1970a. The fine structure of Leucocytozoon
simondi. II. Mega Ioschizoaony. Can. j. Zoo 1. 48:
417-421.
Desser, S.S. 1970b. The fine structure of Leucocytozoon
simondi. III. The ookinete and mature sporozoite.
Can. TT Zool. 48: 641-645.
Desser, S.S. 1972a. Gametocyte maturation, exflagellation
and fertilization in Parahaemoproteus (=Haemoproteus)
v e1 a n s (Coatney and Roudabu s h ) (Haemosporidia:
Haemoprote i dae) an u 1 trastructura 1 study. J.
Protozool. 19: 287-296.

282
Oesser, S.S. 1972b. The fine structure of the ookinete of
Pa r a h a ernop roteus ve 1 an s ( = Ha emop roteus v e 1 a n s )
(Ha emospot idia: Ha emop rotei d a e ) . Ca n. J ^ Zo o I .
50: 477-480.
Desser, S.S. 1972c. The fine structure of Leucocytozoon
simondi. V. The oocyst. Can. J. Zool. 501 707-711.
Des ser, S.S., and F. Allison. 1979. Aspects of the
sporogonic development of Leucocytozoon tawaki of the
Fiordland Crested Penquin in its primary vector,
Austrosimulium u n g u 1 at u m: an u! t rast ructura 1
study. TI Parasitol. 65: 737-744.
Desser, S.S., and A.M. Fa 1 lis. 1967. The cytological
development and encapsulation of mega 1 osch i zon t s of
Leucocytozoon simondi. Can. J. Zool. 45: 1061-1065.
Desser, S.S., A.M. Fallís and P.C.C. Garnham. 1968.
Relapses in ducks chronically infected with
Leucocytozoon simondi and Parahaemoproteus net tionis.
Can. J. Zoo 1. 46: 231-285.
Desser, S.S., and K.A. Wright. 1968. A preliminary study of
the fine structure of the ookinete, oocyst and
sporozoite formation of Leucocytozoon simondi
Mathis and Leger. Can. J. ZooH 303-307.
Dubey, J.P., T.P. Kistner and G. Callis. 1983. Development
of Sarcocyst i s in mule deer transmitted through
dogs and coyotes. Can. J. Zool. 61: 2904-2912.
Duncan, D., J. Eades, S.R. Julian and D. Micks. 1960.
Electron microscopic observations on malarial oocysts
(P1 a smodiurn ca themeriurn). J. Protozool. 7: 18-26.
Dyce, A.L. 1969. The recognition of nulliparous and
parous Cu 1 i c oid e s without dissection. J. Aust.
En tomo 1. Soc. SI TT-15.
Entzeroth, R. 1983. Electron microscope study of
merogony preceding cyst formation of Sarcocystis
sp. in Roe Deer. Z. Parasitenkd. 69: 447-456.
Eve, J.H., F.E. Kellog and R.W. Bailey. 1972. Blood
parasites in wild turkeys of eastern West
Virginia. J. Wildl. Manage. 36: 624-627.

283
Eve, J.H., F.E. Kellog and F.A. Hayes. 1972. Blood
parasitisms of wild turkeys in the southeastern
U.S.A. J. Am. Vet. Med. Assoc. 161: 638-640.
Fallís, A.M., and G.F. Bennett. I960. Description of
Haemoproteus canachites n. sp. (Sporozoa:
Haemoproteidae) and sporogony in Culi coi des (Díptera:
Ceratopogonidae). Can. J. Zool. 38: 455-464.
Fa 1 1 is, A.M., and G.F. Bennett. 1961. Sporogony of
Leucocytozoon and Haemoproteus in simuliids and
ceratopogonids and—a revised classification of
the Haemosporidiida. Can. J. Zool. 39: 215-228.
Fallis, A.M., and G.F. Bennett. 1966. On the epizootiology
of infections caused by Leucocytozoon si mo n d i in
Algonquin Park, Canada. Can~ J~( Zoo 1 . TT": 101-112.
Fallis, A.M., and S.S. Desser. 1977. On species of
Leucocytozoon , Haemop t o t e u s and Hepa t ocy s t i s . I n
"Parasitic Frotozoa" vol. 3. J . P. K r e i e r , ed.
Academic Press, New York. pp. 239-266.
Fallis, A.M., and D.M. Wood. 1957. Biting midges
(Díptera: Ceratopogonidae) as intermediate hosts for
Haemoproteus of ducks. Can. J. Zool. 35: 425-435.
Farmer, J.N. 1965. Gizzard lesions associated with
Ha emo proteus sacharovi infections of pigeons.
Proc. Iowa Acad. Sci. 71: 537-542.
Forrester, D.J., E.C. Greiner and M.K. Kigaye. 1977. Avian
Haemoproteidae. 7. A review of the haemoprot e i ds of
the family Ciconiidae (storks) and descriptions of
Ha emop roteus brodkorbi sp.nov. and Haemoproteus
pe ircei sp.nov. Can. J. Zool. 55: 1268-127TI
Forrester, D.J., L.T. Hon, L.E. Williams Jr. and
D.H. Austin. 1974. Blood protozoa of wild
turkeys in Florida. J. Protozool. 21: 494-497.
Frenkel, J.K. 1973. Toxoplasmosis: parasite life cycle,
pathology, and immunology, in "The Coccidia", D.M.
Hammond and P.L. Long, eds. University Park Press,
Baltimore. 482 pp.
Frenkel, J.K., and J.P. Dubey. 1975. Hammond i a hamnondi
gen.nov., sp.nov., from domestic cats, a new coccidian
related to Toxoplasma and Sarcocystis. Z. Parasitenk.
46: 3-12.

284
Freund, R.J., and R.C. Lit tell, 1981. SAS for linear
models. SAS Institute Inc., Cary, North Carolina.
231 pp.
Gabaldon, A., and G. Ulloa. 1980. Ho 1oendemicity of
malaria: an avian model. Trans. Roy. Soc. Trop.
Med. Hyg. 74: 501-507.
Gallucci, B.8. 1974a. Fine structure of Haemoproteus
columbae Kruse during macrogametogenesis and
fertilization. J. Protozool. 21: 254-263.
Gallucci, B.B. 1974b. Fine structure of Haemoproteus
co1umbae during differentiation of the ookinet e. T7
Protozool. 21: 264-275.
Gardiner, C.H., H.J. Jenkins and K.S. Mahoney. 1984.
Myositis and death in Bobwhites, Col in u s
virginianus (L. ) , due to hemorrhagic cysts ol a
h a emo sporozoan of undetermined taxonomic status.
J. Wi Id. Dis. 20: 308-318.
Garnham, P.C.C. 1951. The mosquito transmission of
Plasmodiurn inui Ha 1berstaedter and Prowazek, and its
pre-erythrocytic development in the liver of the
Rhesus monkey. Trans. Roy. Soc. Trop. Med. Hyg. 45:
45-52.
Garnham, P.C.C. 1966. Malaria parasites and other
Haemosporidia. Blackwell Scientific Publ., Oxford.
1114 pp.
Garnham, P.C.C. 1973a. Epizootics of Leucocy tozoon
infections in parakeets in England, in "Progress fñ
Protozoology". Univerité de Cler mo n t , C1 e r mo n t ,
France, p. 149.
Garnham, P.C.C. 1973b. Unusual hosts for parasites
under natural and experimental conditions, in
"Proc. Fourth Internat. Cong. Protozool.". Université
de Clermont, Clermont, France, pp. 193-194.
Garnham, P.C.C., R.G. Bird, J.R. Baker, S.S. Desser and
H.M.S. El-Nahal. 1969. Electron microscope studies on
motile stages of malaria parasites VI. The ookinete of
Plasmodium berghei yoe 1 i i and its transformation into
the early oocyst. Trans. Roy. Soc. Trop. Med. Hyg.
63: 187-194.

285
Gonder, R. 1915. On the transmission of Haemoproteus
col umbae. Rep. Dir. Vet. Res. Pretoria 3 & 4: 627-032.
Greiner, E.C. 1971. The comparative life histories of
Haemoproteus sacharov i and Haemoproteus macca 11 urni i n
the Mo u r ning Dove (Zenaida macroura) . PF.D.
Thesis. University of Nebraska, Lincoln. TT5 pp.
Greiner, E.C. 1975. Prevalence and potential vectors of
Haemoproteus in Nebraska mourning doves. J. Wildl.
Dis.1l: 150-156.
Greiner, E.C., E.S. Eveleigh and W.M. Boone. 1978.
Ornithophi I ic Cu 1 ico i des spp. (Díptera:
Ce ratopogon i dae ) from New Brunswick, Canada, and
implications of their involvement in haemoproteid
transmission. J. Med. Ent. 14: 701-704.
Greiner, E.C., and D.J. Forrester. 1980. Haemop r o t eu s
me 1 eag ridis Levine 1961: redescription and
d e v e1opmen t a 1 morphology of the gametocytes in
turkeys. J. Parasitol. 66: 652-658.
Greiner, E.C., and A.A. Kocan. 1977. Leucocytozoon
(Haemospori da; Leucocytozoidae) of the Falcon¡formes.
Can. J. Zool. S5: 761-770.
Haberkorn, A. 1968 . Zur hormonellen Bee inf1ussung von
Haemoproteus-Infectionen. Z. Parasitenk. 31:
108-112.
Hanson, H.C., N.D. Levine, C.W. Kossack, S. Kantor and
L.J. Stannard. 1957. Parasites of the mourning dove
(Zenaida macroura carolinensis) in Illinois. J.
Parasitol. 43: 186-193.
Hayes, R.O. 1953. Determination of a physiological saline
solution for Aedes aegypti. (L. ) . J. Econ. Ent. 46:
624-627.
Herman, C.M. 1938. The epidemiology of malaria in
eastern redwings (Agelaius p. phoeniceus). Am. J.
Hyg. 28: 232-243.
Herman, C.M., and G.F. Bennett. 1976. Use of sentinel ducks
in ep i zooti o 1ogica1 studies of anatid blood protozoa.
Can. J. Zool. 54: 1038-1043.

286
Herman, C.M., W.C. Reeves, H.E. McClure, E.M. French and
W.M. Hainnon 1954. Studies on avian malaria in vectors
and hosts of encephalitis in Kern County, California.
Am. J. Trop. Med. Hyg. 3: 676-695.
Howells, R.E., and E.E. Davies. 1971. Nuclear division
in the oocyst of Plasmodiurn berghei. Ann. Trop. Med.
Parasitol. 65: 451-459.
Huff, C.G. 1932. Studies on Haemoproteus of mourning
doves. Am. J. Hyg. 16: 618-627^
Huff, C.G. 1942. Schizogony and gametocyte development
in Leucocytozoon si mo n di, and comparisons with
P1 asmodium and Haemoproteus. J. Infect. Dis. 71:
18-32.
Huff, C.G. 1969. Exoerythrocytic stages of avian and
reptilian malarial parasites. Exo. Parasitol. 24:
383-421.
Humason, G.L. 1979. Animal tissue techniques. W.H.
Freeman and Co., San Francisco. 661 pp.
Janovy, J.,Jr. 1966. Epidemiology of P1 a smodiurn hexamerium
Huff, 1 935 , in Meadowlarks and S t a r 1 i n g~i oT the
Cheyenne Bottoms, Barton County, Kansas. J.
Parasitol. 52: 573-578.
Julian, R.J., T.J. Beveridge and D.E. Galt. 1985. Muscovy
duck mortality not caused by Haemoproteus. J.
Wildl. Dis. 21: 335-337.
Julian, R.J., and D.E. Galt. 1980. Mortality in muscovy
ducks (Cajrina mosciuta) caused by Haemoproteus
infection^ J. Wildl. Dis. 16: 39-44.
Kachigan, S.K. 1982. Multivariate Statistical Analysis.
Radius Press, New York. 297 pp.
Kaye, D., J.G. Merselis and E.W. Hook. 1965. Influence of
P1 a smodium b e rgh ei infection on susceptibility to
s a 1mon e11 a infect ion. Proc. Soc. Exp. Biol. Med.
120: 810-813.
Kellog, F.E., A.K. Prestwood, R.R. Gerrish and G.L.
Doster. 1969. Wild turkey ectoparasites collected in
the southeastern United States. J. Med. Ent. 6:
329-330.

287
Kemp, R.L. 1978. Haemoproteus â–  In "Diseases of Poultry".
M.S. Hofstad, ed. Towa State University Press, Ames,
Iowa. pp. 824-825.
Kettle, D.S. 1965. Biting ceratopogonids as vectors of
human and animal diseases. Acta Tropica. 22: 356-362.
Kettle, D.S. 1968a. The biting habits of Culicoides furens
(Poey) and barbosai Wirth and 31 an ton. Ti The 24-h
cycle, with a note on differences between collectors.
Bull. Ent. Res. 59: 21-31 .
Kettle, D.S. 1968b. The biting habits of Cu 1 icoi des furens
(Poey) and C. barbosai Wirth and Blanton. I I. Effect
of me t eo r o Tog i ca 1 conditions. Bull. Ent. Res. 59:
241-258.
Kettle, D.S. 1977. Biology and bionomics of bloodsucking
ceratopogonids. Ann. Rev. Entomol . 22: 33-51 .
Khan, R.A., S.S. Desser and A.M. Fa Mis. 1969. Survival of
sporozoites of Leucocytozoon in birds for 11 days.
Can. J. Zool. 47: 347-350.
Khan, R.A., and A.M. Fallís. 1969. Endogenous stages of
Parahaemoproteus f r i n g i 1 1 a e (Labbé, 1894 ) and
Leucocytozoon fr ingi 11inarurn Woodcock, 1910. Can. J.
Zool. 47:
Khan, R.A., and A.M. Fallís. 1970. Life cycles of
Leucocytozoon dub r eu i 1 i Mathis and Leger, 1911 and
L e ucocyto z~o o~n fringillinarum Woodcock 1910
(Haemospor i d i a : Leucocytozoidae) . J. Protozool.
17: 642-658.
Khan, R.A., and A.M. Fallis. 1971. A note on the sporogony
of Par ahaemop ro t eu s ve 1 an s (=Haemopro t eu s ve 1 ans
Co a tney and Roudabush) (Haerno s po r i d i a : Haemopro te i dae )
in species of Cu 1 icoi des. Can. J. Zool. 49: 420-421.
Klei, T.R. 1972. The fine structure of Haemoproteus
columbae sporozoites. J. Protozool. 19: 281-286.
Klei, T.R., and D.L. DeGiusti. 1973. U11rastruc tur a 1
changes in salivary glands of Pseudo 1ynchia
canariensis (Díptera: Hippoboscidae) infected with
sporozoites of Haemoproteus columbae. J. Invert.
Pathol. 22: 321-T2ST

238
Klei, T.R., and D.L. DeGiusti. 1975. Seasonal occurrence
of Haemoproteus co1umbae Kruse and its vector
Pseudo 1ynch i a canariensis 3equaert. J. Wild. Dis. 11:
m-nr:
Kozicky, E.L. 1948. Some protozoan parasites of the
eastern wild turkey in Pennsylvania. J. Wildl.
Manage. 12: 263-266.
Kubai, D.F. 1975. The evolution of the mitotic spindle.
Int. Rev. Cyt. 43: 167-227.
Lastra, J., and G.R. Coatney. 1950. Transmission of
Haemoproteus co1umbae by blood inoculation and tissue
transplants. J. Nat 1. Malar. Soc. 9: 151-152.
Leek, R.G., R. Fayer and A. J . Johnson. 1977. Sheep
experimentally infected with Sarcocystis from
dogs. I. Disease in young lambs. ~J~. Paras i to 1 . 63:
642-650.
Levine, N.D. 1961. Protozoan parasites of domestic animals
and man. Burgess Pub. Co., Minneapolis. 412 pp.
Levine, N.D., P.D. Beamer and J. Simon. 1970. A disease of
chickens associated with Arthrocystis ga 11i n.g.,
n. sp., and organism of uncertain taxonomic position.
H.D. Srivastava Comnem. Vol., pp. 429-434.
Levine, N.D., and G.R. Campbell. 1971. A checklist of
the species of the genus Ha emoproteus
(Ap i comp 1exa , P1 asmodiidae). J. Protozool . 18:
475-484.
Lewis, J.C. 1964. Populations of wild turkeys in relation
to fields. Proc. Ann. Conf. S.E. Assoc. Fish &
Wildl. Agencies. 18: 49-56.
Lucas, A.M., and C. Jamroz. 1961. Atlas of avian
hematology. U.S. Dept, of Agriculture, Washington,
D.C. 271 pp.
Lund, E.E., and M.M. Farr. 1965. Protozoa. In "Diseases of
Poultry". H.E. Biester and L.H. Schwarte, eds . Iowa
State Univ. Press, Ames, Iowa. 1382 pp.
Mahrt, J.L., and R. Fayer. 1975. Hematologic and serologic
changes in calves experimentally infected with
Sarcocys tis fus iformis J. Parasitol. 61: 967-969.

289
Markus, M.B., and J.H. Oosthuizen. 1972. Pathogenicity of
Haemoproteus columbae. Trans. R. Soc. Trop. Med.
Hyg. 66: 186-YW.
Mehlhorn, H., W. Peters and A. Haberkorn. 1980. T'ne
formation of kinetes and oocysts in P1 a smod i um
gal 1 inaceum (Haemospor idia ) and considerations on
phylogenetic relationships between Haemosporidia,
Piroplasmida and other Coccidia. Protistologica. 16:
135-154.
Miller, L.H., M. Aikawa, J. Johnson and T. Shiroishi.
1978. interaction between cytochalasin B-treated
malarial parasites and red cells: attachment and
junction formation. J. Exp. Med. 149: 172-184.
Miller, R.E., D.W. Trampel, S.S. Desser and W.J. Soever.
1983. Leucocytozoon simondi infection in European and
Ame ricam eiders . F! Am. Vet. Med. Assoc. 183:
1241-1244.
Miltgen, F., I. Landau, N. Ratanaworabhan and S. Yenbutra.
1981. Parahaemoproteus desser i n.sp.; gamitogonie et
schiz og o nie chez 1 1 h 6 t e n a t u r e 1 ; Psittacula
ros ea t a de Thailande, et sporogonie expér¡mental e
chez Cu 1icoi des nubeculosus . Ann. Parasitol.
Hum. CompF 56: 123-130.
Mohammed. A.H.H. 1965. Studies on the schizogony of
Haemoproteus columbae Kruse 1890. Proc. Egypt.
Acad . Sc i . 19: TTW.
Morehouse, N.F. 1945. The occurrence of Haemoproteus in
the domesticated turkey. Tran. Am. Microsc. Soc.
64: 109-111.
Mor i i , T. 1972. Presence of antigens and antibodies in
the sera of chickens infected with A k i b a
caul 1eryi. Nat. Inst. Hlth. Quart. 12: 161-167^
Mor i i i T. , S. Kitaoka and K. Akiba. 1965. Some
investigations on the sporogony of Leucocy tozoon
caul 1 er y i in laboratory-reared biting midges oT
tour Cu 1 i c oid e s species. Nat. Inst. Anim. Hlth.
Quart. 5: 109-110.
Mundy, B.L., I.K. Barker and M.D. Rickard. 1975. The
developmental cycle of a species of Sa r cocy s t i s
occurring in dogs and sheep, with observations on the
pathogenicity in the intermediate host. Z.
Parasitenk. 46: 111-123.

290
Nautical Almanac.
D.C.
Nautical Almanac.
D.C.
Nautical Almanac.
D.C.
1982.U.S. Govt. Print. Off., Washington,
1983.U.S. Govt. Print. Off., Washington,
1984.U.S. Govt. Print. Off., Washington,
Nathan, M.B. 1981. A study of the diurnal biting and flight
activity of Cu 1 i coid e s ph1e bo t omu s (Williston)
(Dipt era: Ceratopogonidaej using tTTree trapping
methods. Bull. Ent. Res. 71: 121-128.
Newberne, J.W. 1957. Studies on the h i stopatho 1ogy of
Leucocvtozoon simondi infections. Am. J. Vet. Res.
IT: ”191-199.
Nolan, R.A., A.H. Brush, N. Arnheim and A.C. Wilson.
1975. An inconsistency between protein resemblance and
taxonomic resemblance: immunological comparison of
diverse proteins from gallinaceous birds. Condor 77:
154-159.
Oliver, J.E.
John Wiley
1 973 . CI ima t e
and Sons, Inc.,
and man's environment
New York. 517 pp.
Opitz, H.M., H.J. Jakob, E. Wiensenhuetter and V. Vasandra
Devi. 1982. A myopathy associated with protozoan
schizonts in chickens in commercial farms in
peninsular Malaysia. Avian Path. 11: 527-534.
O'Roke, E.C. 1930. The morphology, transmission, and
life history of Haemoproteus 1ophortyx O'Roke, a blood
parasite of th e Califo r ni a val ley quail. U n i v .
Calif. Publ. Zoo 1. 36: 1-50.
Powell, J.A. 1967. Management of the Florida Turkey
and the Eastern Turkey in Georgia and Alabama. In
"The Wild Turkey and its Management". O.H. Hewitt,
ed. The Wildlife Society, Valley Offset, Inc.,
Deposit, New York. pp. 409-451.
Rogge, D. 1 968 . Experi mente11e Bee inf1ussung der
Ha emo proteus - Parasitaemie beiin Grunfinken
(Ca r d u e1 is c h1 o ris) durch kunstlichen Langtag.
Acta Parasit. Pol. T5: 397-407.
Rudzinska, M.A., and W. Trager. 1968. The fine structure
of trophozoites and gametocytes of P1 a smodium
coatneyi . J. Protozool. 1 5: 73-88 .

291
Russell, P.F. 1959. Insects and the epidemiology of
malaria. Ann. Rev. Ent. 4: 415-434.
Russell, P.F., B.N. Mohan and P. Putnam. 1943. Some
observations on spleen volume in domestic fowls in the
course of P1 a smodiurn gal 1 in a c e urn studies. J.
Parasitol. 2T1 203-216.
SAS User's Guide: 3asics. 1982. SAS Institute, Cary, North
Carolina. 921 pp.
SAS User's Guide: Statistics. 1982. SAS Institute, Cary,
North Carolina. 584 pp.
Schorger, A.W. 1966. The wild turkey. University of
Oklahoma Press, Norman. 625 pp.
Schrevel, J., G. Asfaux-Foucher and J.M. Bafort. 1977.
Etude u1trastructura1e des mitoses multiples au Cours
de la sporogonie du Plasmodium b. berghei. J.
Ultrast. Res. 59: 332-3STH ~
Sergent, Ed., and Et. Sergent. 1906. Sur le second hote de
I 'Ha emo proteus ( Halteridium) du pigeon. (Note
préliminaire). C.RT Séances Soc. Biol. Fil. 61:
494-496.
Service, M.W. 1971. Adult flight activities of some
British Cu 1icoi des species. J. Med. Ent. 8: 605-609.
Sibley, L.D., and J.K Werner. 1984. Susceptibility of
pekin and muscovy ducks to Haemoproteus nettionis.
J. Wild!. Dis. 20: 108-113.
Sinden, R.E., and K. Strong. 1973. An u11rastruetura 1 study
of the sporogonie development of P1 a smodi urn f a 1 c ipa r urn
in Anopheles gambiae. Trans. Roy. Soc. Troo. Med.
Hyg. 72: 477-491.
Smith, H.A. , T.C. Jones and R.D. Hunt. 1972. Veterinary
Pathology. Lee and Febiger, Philadelphia. 1521 pp.
Snow, W.E. 1955. Feeding activities of some
b 1 ood-sucking Díptera with reference to vertical
distribution in bottomland forest. Ann. Ent. Soc.
Amer. 48: 512-521.
Sterling, C.R. 1972. U11rastructura 1 study of gametocytes
and gametogen esis of Haemoproteus me t chn i kov i . J.
Protozool. 19: 69-76.

292
Sterling, C.R., and M. Aikawa. 1973. A comparative study
of gametocyte ultrastructure in avian Haemospor i d i a .
J. Protozool. 20: 81-92.
Sterling, C.R., and D.L. DeGiusti. 1972. Ultrastructura 1
aspects of schizogony, mature sc'nizonts and merozoites
of Haemoproteus metchnikovi. J. Parasitol. 58:
641-652.
Sterling, C.R., and D.L. DeGiusti. 1974. Fine structure of
differentiating oocysts and mature sporozoites of
Haemoproteu s metchnikovi in its intermediate host
Cli r y sops ca 1 1 i dus â–  J. Protozool. 21: 276-283.
Stoddard, H.L. 1963. Maintenance and increase of the
eastern wild turkey on private lands of the coastal
plain of the deep southeast. Bull. Tall Timbers
Res. Station. 3:
Tanner, G.D., and E.C. Turner, Jr. 1974. Vertical
activities and host preferences of several Cu 1 icoi des
species in a southwestern Virginia forest. Mosq. News
34: 66-70.
Tarshis, l.B. 1955. Transmission of Haemoproteus 1ophortyx
O'Roke of the California quail by hippoboscid flies of
the species Sti 1bometopa impressa (Bigot) and Lynchia
hirsuta Ferris. Exp. Paras i tol . T": 464-492.
Taliaferro, W.H. 1941. I mm unology of the parasitic
protozoa, in "Protozoa in Biological Research".
G.N. Calkins and F.M. Summers, eds. University
Press, New York. 830-854.
Terzakis, J.A. 1971. Transformation of the P1 a smod i uin
cynomolgi oocyst. J. Protozool. 18: 62-73.
Terzakis, J.A., H. Sprinz and R.A. Ward. 1966. Sporoblast
and sporozoite formation in P1 a smodiurn gal Iinaceum
infections of Aedes aegypt i . Mi 1. Med. F3 1: 984-992.
Terzakis, J.A., H. Sprinz and R.A. Ward. 1967. The
transformation of the PI asmodiurn gal Iinaceum oocyst in
Aedes aegypti mosquitoes. J. Cell. 3io1. TT: 311-326.
Trefiak, W.D., and S.S. Desser. 1973. Crystalloid
inclusions in species of Leu cocytozoon ,
Parahaemoproteus and Plasmodium. T! Protozool. ZD:
73-80.

293
Vanderberg, J.P. 1977. PI a ¡modium b e r g h e i :
quantitation of sporozoites injected by mosquitoes
feeding on a rodent host. Exp. Parasitol. 42:
169-181 .
Viens, P. , A. Tarzaali and M. Quevillon. 1974. Inhibition
of the inmune response to pertussis vaccine during
Plasmodium berghei infection in mice. Am. J.
Trop. ivied. Hyg. 23: 346-849.
Wallace, G.D. 1973. Sa rcocys tis in mice inoculated with
Toxop1 asma-1ike oocysts from cat feces. Science 180:
1375-1377.
von Wasielewski, T., and G. WUlker. 1918. Die Hdmoproteus
Infection des TUrmfalken. Arch. Schiffs. Tropenhyg.
22: 1-100.
Wenyon, C.M. 1926. Protozoology. 2 vols. Bailliere, Tindall
and Cox, London. 1563 pp.
Wernsdorfer, W.H. 1980. The importance of malaria in the
world. In “Malaria" vol. 1. J.P. Kreier ed. Academic
Press, New York. pp. 1-93.
White, E.M., and G.F. Bennett. 1979. Avian
haemoprot e i dae. 12. The haemoprote i ds of the grouse
family Tetraonidae. Can. J. Zool. 57: 1465-1472.
Williams, J.E. 1978. Paratyphoid infections. In "Diseases
of Poultry". M.S. Hofstad, ed. Iowa State
University Press, Ames, Iowa. pp. 117-167.
Wong, T.C., and S.S. Desser. 1976. Fine structure of oocyst
transformation and the sporozoites of Leucocytozoon
du'oreu i 1 i â–  J. Protozool. 23: 115-126.

BIOGRAPHICAL SKETCH
Carter Tait Atkinson was born on November 23, 1954, in
Bethesda, Maryland. He graduated from Gaithersburg High
School in June, 1972, and received a Bachelor of Science
degree with honors in biology from Dickinson College in
May, 1976. During the following 3 summers, he worked as a
field ornithologist for the U.S. Fish and Wildlife Service
on their Hawaii Forest Bird Survey. While working in
the Hawaiian Islands, he became interested in the role
of introduced diseases in the decline and extinction of
native Hawaiian birds. This interest led him to enroll as
a graduate student at the University of Maryland between
September, 1978, and May, 1979. In August, 1979, he
enrolled as a student in the Department of Tropical
Medicine and Medical Parasitology at the Louisiana State
University Medical Center in New Orleans. He graduated in
August, 1981, and received a Master of Science degree in
parasitology. In September, 1981, he enrolled in the
graduate program at the University of Florida.
294

1 certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi Iosophy.
Donald J. Forrester, Chairman
Professor of Veterinary Medicine
1 certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi 1osophy.
fTT i s
Associate
Med icine
ire i ner ,' Cocha i rman
Professor of Veterinary
I certify that 1 have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi 1osophy.
Assistant Professor of Entomology
and Nema to logy

I certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Philosophy.
Emeritus Professor of Veterinary
Med icine
I certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Philosophy.
Martin D
.7^
/ /
y/-.
Voung /j
Professor
jLI
Research
Medicine
of Veterinary
This dissertation was submitted to the Graduate Faculty of
the College of Medicine and was accepted as partial
fulfillment of the requirements for the degree of Doctor of
Philosophy.
De c emb e r, 1985
Dean, Graduate School

UNIVERSITY OF FLOI^
!| !l IÜÜ4 6413



4000
3500
3300
2500
2000
1500
1000
500
0
WEEKS
132


Figure 40. Average plasma protein concentrations for high dose
birds (O O ), low dose birds (A A ) and control
birds (# ). Statistically significant differences
among groups for each week of the study are indicated by
the boxed points (p 0.05).


238
Klei, T.R., and D.L. DeGiusti. 1975. Seasonal occurrence
of Haemoproteus co1umbae Kruse and its vector
Pseudo 1ynch i a canariensis 3equaert. J. Wild. Dis. 11:
THTT3T
Kozicky, E.L. 1948. Some protozoan parasites of the
eastern wild turkey in Pennsylvania. J. Wildl.
Manage. 12: 263-266.
Kubai, D.F. 1975. The evolution of the mitotic spindle.
Int. Rev. Cyt. 43: 167-227.
Lastra, J., and G.R. Coatney. 1950. Transmission of
Haemoproteus co1umbae by blood inoculation and tissue
transplants. J. Nat 1. Malar. Soc. 9: 151-152.
Leek, R.G., R. Fayer and A. J Johnson. 1977. Sheep
experimentally infected with Sarcocystis from
dogs. I. Disease in young lambs. ~J~. Parasito 1. 63:
642-650.
Levine, N.D. 1961. Protozoan parasites of domestic animals
and man. Burgess Pub. Co., Minneapolis. 412 pp.
Levine, N.D., P.D. Beamer and J. Simon. 1970. A disease of
chickens associated with Arthrocystis ga 11 i n g ,
n. so., and organism of uncertain taxonomic position.
H.D. Srivastava Comnem. Vol., pp. 429-434.
Levine, N.D., and G.R. Campbell. 1971. A checklist of
the species of the genus Ha emoproteus
(Ap i comp 1exa P1 asmodiidae). J. Protozool 18:
475-484.
Lewis, J.C. 1964. Populations of wild turkeys in relation
to fields. Proc. Ann. Conf. S.E. Assoc. Fish &
Wildl. Agencies. 18: 49-56.
Lucas, A.M., and C. Jamroz. 1961. Atlas of avian
hematology. U.S. Dept, of Agriculture, Washington,
D.C. 271 pp.
Lund, E.E., and M.M. Farr. 1965. Protozoa. In "Diseases of
Poultry". H.E. Biester and L.H. Schwarte, eds Iowa
State Univ. Press, Ames, Iowa. 1382 pp.
Mahrt, J.L., and R. Fayer. 1975. Hematologic and serologic
changes in calves experimentally infected with
Sarcocys tis fus iformis J. Parasitol. 61: 967-969.


studies in elucidating the taxonomic and phylogenetic
relationships among the Haemosporina, only 1 study oE
the oocysts of Haemoproteus has been published (Sterling
and DeGiusti, 1974). These workers found that H.
me t c h nik o v i a chelonian parasite that develops in deer
flies, shares features of its sporogonic development
w i t h P 1 a smod i urn and Leucocy t o zoo n that neither of
these2o genera hold in comuon.
Differentiation of the oocyst. The sporogony of avian
and mammalian species of P1 a smod i urn follows a number of
similar steps (Mehlhorn et al., 1980; Sinden and Strong,
1978; Garnhan et al., 1969; Terzakis et al., 1967; Terzakis
et al., 1966; Duncan et al., 1960). After penetrating the
midgut, the ookinete rounds-up under the basement
membrane. Mehlhorn et al. (1980) found the remains of
polar rings and mictonemes derived from the apical complex
of the ookinete at the periphery of early P^ ga 1 1 inaceum
oocysts. Similarly, Garnham et al. (1969) found
remnants of this complex in a dimple or small groove in the
peripheral cytoplasm of early oocysts of P^ b e r g h ei
yoe1 ii. Studies of older PI asmodium oocysts have failed to
find evidence that these organelles persist more than a few
days. Among other Haemosporina, the remnants of apical
organelles from the ookinete have been observed in
early, but not later oocysts of t awaki (Des ser and


Figure 41. Average hemoglobin concentrations for high dose birds
(O O ), low dose birds ( A A ) and control birds
( @ ). Statistically significant differences among
groups for each week of the study are indicated by
the boxed points (p 0.05).


21
cage to Cu 1icoides-proof turkey rooms and held for 4 weeks
to allow any infections acquired in the field to become
patent. They were replaced in the field on the same day
with unexposed, 2-week-old poults. The sentinel birds
were bled from a leg vein, 3 times a week for 4 weeks,
following their exposure in the field. Blood smears were
fixed with absolute methanol and stained with 10% Giemsa,
pH 7.2. Infections were diagnosed by scanning
approximately 10,000 red blood cells at 1000X. All turkeys
used in this study were obtained as day-old poults from
Thaxton's Turkeys (P.O. Box 127, Watki nsvi 11e, Georgia).
Vectors
Once every 2 weeks a Bennett trap (Bennett, 1960)
and a New Jersey light trap were operated at each site
within 50 m of the sentinel cages. The Bennett trap was
operated in the middle level of the forest canopy where
ornithoph i 1 i c species of Cu 1 icoi des are most active (Tanner
and Turner, 1974) from approximately I hour before sunset
to 1 hour after sunset. A turkey was placed into an 0.3
cubic meter welded wire cage made from 1.3 cm mesh and
hoisted on an 0.6 square meter plywood board into the
canopy by means of lightweight nylon rope and small
pulleys. Following an exposure of 10 20 minutes, the
bird was lowered quickly to the ground and covered with
an 0.6 cubic meter wooden frame screened with fine nylon


10
The infected birds tended to have lighter body weights,
enlarged, blackened spleens and livers that were slightly
smaller than normal. O'Roke (1930) felt that the
parasitized blood cells had less oxygen carrying capacity
and were less elastic and more likely to rupture when
passing through small capillaries. He described 4 stages
of disease in birds infected with fU lophortyx: 1)
mild-chronic with no obvious signs of infection, 2)
mild-acute where birds were restless and had poor appetites
for 2 to 4 days, 3) moderate-chronic where birds were
anemic, weak and more susceptible to death by exposure
and exhaustion and 4) heavy-acute where birds lost weight,
refused food, were unable to fly and eventually died.
O'Roke commonly observed birds with moderate-chronic
infections in the field, but saw only 4 heavy-acute
infections in the several hundred birds he examined.
Unfortunately, O'Roke did not attempt to reproduce the
disease in experimentally infected quail or precisely
quantify parasitemias and course of infection as they
related to the stage of the disease.
More recently, Julian and Galt (1980) reported several
incidents of a pathogenic Haemoproteus infection in Muscovy
Ducks, Cairina mo s c h a t a from Ontario. They found large
numbers of schizonts, morphologically similar to those
of other species of Haemop r o t eu s in endothelial cells
from a variety of tissues. The schizonts appeared to


113
areas of diffuse, white streaks, several rnn in length,
embedded deeply in the tissue. These were less distinctive
than the white fascia that separated muscle bundles. Fully
mature megaloschizonts containing numerous, densely packed,
spherical zoites and several large, central vacuoles
were present in sections of the tissue (Figure 50, 51).
Immature forms resembling 14-day-old schizonts were also
present. The mega losch i zon t s were surrounded by a thick
hyaline wall and ranged from 30 to 113 urn in diameter.
They extended as far as 465 urn along the long axis of
muscle fibers. Fibers adjacent to the mega 1oschizonts were
swollen, pale and hyaline (Figure 50). Several layers
of connective tissue, infiltrated with macrophages,
surrounded some schizonts. A large degenerating
meg a 1 oschizont filled with amorphous, gray material
containing irregularly shaped red masses, was present in 1
section. The mega 1oschizont was surrounded by giant cells
and an outer layer of connective tissue (Figure 56).
Ruptured, partially empty mega 1oschizonts that contained
small numbers of scattered zoites were also present (Figure
52). Each zoite contained a small mass of chromatin and a
large vacuole. Giemsa-sta i ned erythrocytes from the
same bird contained young gametocytes that were
morphologically indistinguishable from the exoerythrocytic
zoites.


166
Table 20. Classification sumnaty of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with macrogarnetocy tes The discriminant
function derived from data in Table 18 was used
to classify the discriminant scores.
Classified Into Species
Species
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
1 (25.0)+
0 (0.0)
3 (75.0)
0
(0.0)
4
(100)
Pheasant
2 (66.7)
0 (0.0)
1 (33.3)
0
(0.0)
3
(100)
Turkey
1 (25.0)
0 (0.0)
2 (50.0)
1
(25.0)
4
(100)
Total
4 (36.4)
0 (0.0)
6 (54.6)
1
(9.1)
11
(100)
Priors*
0.4054
0.1892
0.4054
+ Percentage of total
* Prior probability of being assigned to that class


Figure 57. Mega 1 oschizonts from pectoral muscle of a
naturally infected Wild Turkey collected
near Lake Apopka, Orange County, Florida in
November, 1970. Mega loscli i zont s have a thick,
hyaline outer wall (arrows). Hematoxylin
and eosin. Bar = 50 um.
Figure S8. Mega 1oschizont from pectoral muscle of a
naturally infected Wild Turkey collected
near Lake Apopka, Orange County, Florida in
November, 1970. Mega 1 oschizont has a thick,
hyaline wall (arrow) and is surrounded by pale,
swollen and hyaline muscle fibers (M).
Hematoxylin and eosin. Bar = 50 um.


Figure 67. Circulating macrogametocyte. The parasite
is bound by a 3-layered pellicle (arrow) and
has a branched nucleus (N) with a prominent
nucleolus (Nu), mitochondria (M) with
tubular cristae and osmiophilic bodies (double
arrow). The endoplasmic reticulum (Er)
contains a moderately dense, amorphous
material. X 46,400.
Figure 68. Higher magnification of Figure 67. The
endoplasmic reticulum (Er) is continuous
with the inner, osmiophilic layer of the
pellicle (arrow). X 72,000.


igure 17. Average monthly temperatures at Paynes Prairie
( -ft ) and Fish eating Creek ( O O )
during 1982, 1983 and 1984 (Climatological
Data: Florida, 1982, 1983, 1984).


MONTHS
INCHES INCHES INCHES


182
66


LOG |o (N +1) LOG |Q ( N +1) LOG,q(N + I)
77
Culicoides edeni


18
309 eastern Wild Turkeys collected in the southeastern
U.S., Kellog et al. (1969) only found hippoboscid flies
occasionally. Forrester (unpublished) has never recovered
hippoboscids from Florida Wild Turkeys. Despite the
uncommon occurrence of hippoboscid flies on Wild Turkeys,
the prevalence of Ha einop roteus infections is high.
Forrester et al. (1974) found 69% of 85 Wild Turkeys from
northern Florida and 87% of 399 Wild Turkeys from southern
Florida infected with this parasite. Other workers have
found prevalences ranging from 5% in Pennsylvania (Kozicky,
1948) to 80% in southern Texas (Cook, et al., 1966).
The rarity of hippoboscid flies, the high prevalence of
Haemop r ot eus infections and the fact that transmission
of the parasite to caged domestic turkeys readily occurs
when they are placed in a suitable habitat (Forrester,
et al., 1974) suggest that ceratopogonids are the primary
vectors of fC me 1eag rid i s in Florida. In confirmation
of this, Atkinson et al. (1983) recently demonstrated
that at least 3 species of Florida Cu 1 icoi des could support
complete development of the sporogonic stages of H,
me 1eag ridis.
This study was undertaken to
(1) determine the vectors and investigate the seasonal
transmission and ep i zooti o 1ogy of me 1eag rid i s in
Florida,


257
on the basis of measurements alone. All 3 species were
retained because experimental studies had demonstrated
their specificity to particular host families. In view of
the recent revisions in avian taxonomy, similar studies of
host specificity of avian haemoproteids, e g H
me 1eag ridis and man soni, should be conducted before
synonymies are proposed.
Fine Structure
Mature Gametocytes
Mature gametocytes of me 1 eagridis were similar
in fine structure to gametocytes of other species of
Haemoproteus Ultrastructural features of the
cytostoine, nucleus and nucleolus, mitochondria, osmiophilic
bodies, cytoplasmic ribosomes and endoplasmic reticulum
were essentially identical to gametocytes of H.
co 1 umba e v e 1 an s and me t c h n i k o v i (Bradbury and
Roberts, 1970; Sterling, 1972; Sterling and Aikawa,
1973; Bradbury and Trager, 1968a; Desser, 1972a).
Differences from other species of Haemoproteus were
apparent in the pellicle of meleagridis. Studies of
gametocytes of other species have shown that the
pellicle is composed of 3 layers 2 outer unit


251
Mac rosametogen es is Few detailed ultrastructural
observations of the maturation of ma ergame tes have been
made for any species of Haemoproteus 3radbury and Trager
(1968a) provided the most detailed description. During the
initial steps of gametocyte maturation in columbae, they
noted the detachment of the endoplasmic reticulum from the
nuclear envelop and the elongation of the
macrogametocyte nucleus from a spherical to an ellipsoidal
shape. The changes in nuclear shape were accompanied by
the appearance of an intranuclear spindle that converged on
an electron dense plaque in the nuclear membrane. Two
"atypical" centrioles composed of a circle of 9, single
microtubules rather than the conventional circle of 9,
triple microtubules were embedded in an electron dense
material. They appeared in the cytoplasm next to the
electron dense plaque. Approximately 10 minutes after the
start of gametocyte maturation, the nucleus constricted
twice to form 2, detached maturation bodies. The
intranuclear spindle as well as the nucleolus were retained
by the nucleus.
A similar process appeared to occur in maturing
inacrogametocytes of H_j_ meleagridis although
observations are limited. Unfortunately, gametocyte
maturation was not followed longer than 3 minutes, so
the formation of maturation bodies was not observed.
Atypical centrioles were not found in association with the


Figure 75.
Figure 76.
Extracellular maturing rnacrogamete. The gamete
has a pellicle composed of 2 continuous layers
(arrow), an elongate nucleus (N) with a
nucleolus (Nu), mitochondria (M) and an
extensive network of endoplasmic reticulum
(small arrows). X 34,800.
A higher magnification of a portion of
Figure 75. The macrogamete has a cytostome
(large arrow) that is surrounded by 2 electron
dense thickenings. The pellicle is composed of
2 uninterrupted unit membranes (small
arrows). X 74,750.


103
Focal aceas of enteritis characterized by the presence
of heterophils in the lamina propria and submucosa were
present in sections of intestine and cecum. Coccidian
parasites were not detected. Sections of kidney, brain,
bone marrow, proventri cu 1 us and gizzard were unremarkable.
Microscopic observations surviving birds. At 8
weeks post-infection, nodular infiltrates of mononuclear
cells, macrophages, heterophils and giant cells were
evident in sections of skeletal muscle from low dose and
high dose birds (Figure 33). The hyaline remnants of
the outer wall of degenerating mega 1oschizonts and necrotic
and calcified muscle fibers were at the center of some
of the nodules. A scattered lymphocytic, heterophilic
infiltrate was frequently perivascular and often present
between muscle fibers. Pectoral muscle from 1 low dose
bird contained a degenerating cyst with merozoites (Figure
34). Thrombi surrounded by macrophages and heterophils
were present in some sections (Figure 34). Remnants of
degenerating muscle fibers, infiltrated with macrophages
and heterophils, were scattered randomly throughout the
sections of muscle (Figure 35).
Sections of liver, lung and spleen from both low
and high dose birds contained moderate to extensive, random
deposits of pigment. No pigment was found in control
birds. Deposits in the liver and spleen were massive


31
to determine the approximate dosage. Each of the 12 birds
was inoculated with 0.5 cc of slurry containing
approximately 4,400 sporozoites.
Birds in the second experimental group were inoculated
IP with separate pools of 35 specimens of edeni ground
and quantified as above. Each bird was inoculated with
0.5 cc of slurry containing approximately 57,500
sporozoites. Birds in the control group were inoculated
IP with 0.5 cc of RPM1 tissue culture fluid containing
10% turkey serum.
Following inoculation the birds were housed in groups
of 3 in 12 battery cages in a vector-proof room. Each
compartment held 1 bird from each experimental group.
Birds were assigned to the compartments with a random
number table. The poults were fed and watered as described
earlier.
Twenty-four hours prior to their inoculation (Week
0), and once a week for 8 weeks following infection, each
bird was weighed and the tarsometatarsal length of the
right leg was measured with calipers. Two heparinized
capillary tubes were filled with blood from a wing vein.
Blood was immediately drawn from I capillary tube into
2, 20 ul pipettes. The contents of each pipette was
irrmediately dispensed into each of 2 separate test tubes
containing 5 mis of cyanomethemag1ob i n reagent (1:251
Cyanomethemag 1 ob i n Test Kit, Fisher Scientific). The


169
Table 23. Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of host cells infected
with microgametocytes. The discriminant
function derived from data in Table 21 was
used to classify the discriminant scores.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
2
(50.0)+
1
(25.0)
0
(0.0)
1 (25.0)
4
(100)
Pheasant
0
(0.0)
4
(100)
0
(0.0)
0 (0.0)
4
(100)
Turkey
1
(25.0)
1
(25.0)
1
(25.0)
1 (25.0)
4
(100)
Total
3
(25.0)
6
(50.0)
1
(8.3)
2 (16.7)
12
(100)
Priors*
0.
.3333
0.
,3333
0.
,3333
+ Standard Deviation
* Prior probability of being assigned to that class


251
large number o£ often poorly described species in this
genus, it is totally dependent on the maintenance of a
stable system of classification of the avian hosts.
Recently, the Committee on Classification and Nomenclature
of the American Ornithologist's Union (1983) published a
revised sixth edition of the Checklist of North American
Birds For the first time since the publication of the
third edition in 1910, the committee incorporated major
changes in the systematics of higher categories. These
changes were based on the conservative evaluation of
many new discoveries in the morphology, paleontology,
biochemistry, genetics and behavior of avian species
(Checklist of North American Birds, 1983). While the
status of many orders and families remained unchanged,
extensive revisions were made in others, e.g.
Passeriformes, Gall¡formes, that may require corresponding
revisions in haemoproteid taxonomy.
Prior to the 1983 revision of the Checklist of
North American Birds, the Order Gall¡formes was composed of
5 families: Cracidae (Curassows and Guans), Phasianidae
(Pheasants and Quail), Tetraonidae (Grouse), Me 1eagrididae
(Turkeys) and Numididae (Guineafowl). In the 1983 edition
of the Checklist of No r t h Ame rican Birds, only 2
families were recognized in the Order Gall¡formes: Cracidae
(Curassows and Guans) and Phasianidae (Pheasants, Quail,
Grouse, Turkeys, Guineafowl). The families Phasianidae,




247
Leucocytozoon have documented increases in parasite
specific immunoglobulins as the infections progressed
(Congdon et al., 1969; Mor ii, 1972).
3y o weeks pos t i n f ec t i on when the experiment was
ended, surviving birds had regenerating muscle fibers
and scarring associated with the destruction and removal of
mega 1 oschizonts (Figures 33 34 35 ). The persistence
of some meg a 1 oschizonts up to 8 weeks post-infection
suggests that they may be a source for relapses.
It is impossible to determine whether the high
(33%) mortality in the high dose group resulted from the H.
me 1eag rid i s infection alone, or whether the concurrent
SalmoneI 1 a infection was a significant factor. Studies of
the interactions between P^ be r g h ei and S a Imone11 a
typhimuriurn have shown that mice infected with both agents
died earlier than mice infected with either agent alone
(Kaye et al., 1965). Viens et al. (1974) had similar
results when they infected mice with P_^ y o e 1 i i and
Bo r de t e1 la pe r t u s s i Cox (1978 ) suggested that the
synergism between Plasmodium infections and other
infectious agents may result from the imnunodepression that
often accompanies P1 a smodiurn infections. Similar
studies have not been conducted with Haemoproteus.
Because of the way they were housed, all birds in each
of the 3 experimental groups had an equal exposure to
Salmone11 a enteriditis. The close proximity of the battery


269
subdivision of the oocyst cytoplasm into many sporoblast
bodies and the endogenous or internal sporozoite budding
that occurs in P1 a smod i urn is probably an adaptation to
increase the surface area available for sporozoite
formation (Sinden and Strong, 1978).
Limited observations of me 1 eagr i d i s from this study
suggest that vacuolization and cleft formation do not
occur. Thus, this parasite closely resembles Leucocytozoon
in size of oocysts and number of sporozoites produced as
well as in contraction of the sporoblast body and
differentiation of the sporozoites.
Nuclear divisions. The nuclear events following
fertilization of haemosporidian parasites are still poorly
understood. Bano (1959) and Canning and Anwar (1963) found
evidence of post-zygotic meiosis in the early oocysts of
P1 a smodi urn. They observed what they believed to be 4
chromosomes in diploid oocysts and 2 in haploid oocysts in
stained preparations. Later studies with electron
microscopy have shown that the chromosomes do not
condense. The dark-sta in i ng nuclear masses observed
earlier may have been nucleoli (Mehlhorn et al., 1980)
or fragments of an enlarged digitate nucleus (Canning
and Sinden, 1973). Mehlhorn et al. (1980) observed nuclear
spindles in zygotes and in ookinetes of P^ ga 1 1 i naceum and


Figure 28. A degenerating mega 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-in£ection .
The hyaline cyst wall (arrow) and a mixed
inflammatory infiltrate composed primarily
of red blood cells is all that remains of
the schizont. Hematoxylin and eosin. Bar = SO
um.
Figure 29. A venule blocked by a thrombus in pectoral
muscle of a high dose bird that died
spontaneously at 22 days post-in£ection .
The thrombus is surrounded by a giant cell
(arrow). Hematoxylin and eosin. Bar = 50 um.


BIOGRAPHICAL SKETCH
Carter Tait Atkinson was born on November 23, 1954, in
Betbesda, Maryland. He graduated from Gaithersburg High
School in June, 1972, and received a Bachelor of Science
degree with honors in biology from Dickinson College in
May, 1976. During the following 3 summers, he worked as a
field ornithologist for the U.S. Fish and Wildlife Service
on their Hawaii Forest Bird Survey. While working in
the Hawaiian Islands, he became interested in the role
of introduced diseases in the decline and extinction of
native Hawaiian birds. This interest led him to enroll as
a graduate student at the University of Maryland between
September, 1978, and May, 1979. In August, 1979, he
enrolled as a student in the Department of Tropical
Medicine and Medical Parasitology at the Louisiana State
University Medical Center in New Orleans. He graduated in
August, 1981, and received a Master of Science degree in
parasitology. In September, 1981, he enrolled in the
graduate program at the University of Florida.
294


49
New Jersey trap catches. Cu 1 icoi de s e deni and
Cu 1 i co i des h i n in a n i were the most common species
collected in the CC>2-baited New Jersey suction traps.
During the March, April and May, 1983, collecting trips to
Fisheating Creek, specimens of eden i had a peak in
activity during the 2-hour sampling period that included
sunset (Figure 11). Individuals of this species were
active at low levels during the night in March and April.
During all 3 collecting trips, activity increased following
sunrise and continued throughout the day at levels lower
than the evening peak. During the day, specimens of C.
eden i were often observed crawling on the head and on
the back feathers of sentinel turkeys exposed on the
ground. Most individuals of edeni were captured in the
suction trap that was operated in the canopy (Figure 11).
During the April and May collecting trips,
specimens of hinmani had peaks in activity during the
2-hour sampling period that included sunset and during the
early morning hours following sunrise (Figure 12).
Activity during the April trip continued into the early
afternoon. Similar diurnal activity did not occur
during the March collecting trip. All specimens of C.
hinmani were captured in the canopy trap. This species
was never captured in Bennett traps operated on the ground.


249
effects in natural Haemoproteus infections have done little
to refute this view. The best documented pathology in
natural Haemoproteas infections was done by O'Roke
(1930) with his study of lophor t yx in California Valley
Quail. He attributed the morbidity and mortality he
observed to anemia caused by rupture of parasitized
erythrocytes and felt that the gametocytes made their host
cells more fragile and susceptible to rupture as they
passed through the capillary beds. The results of this
study indicate that me 1 e ag r i d i s can be severely
pathogenic in high doses and can have detectable effects on
growth and weight at low doses. The pathogenic effects
associated with the experimental infections was more
similar to host reactions to megaloschizonts of
L e ucocy tozoon than to the erythrocyte destruction and
anemia caused by ?1asmodiurn.
Since the lesions in the naturally infected Wild
Turkey found by Nair and Forrester were similar to those of
the experimentally infected birds, 1C me 1eagridis may be a
cause of morbidity and mortality in Wild Turkeys. It
may only be significant in holoendemic areas such as
southern Florida and perhaps southern Texas where the
prevalence of the parasite ranges from 90-100% and rates of
transmission are high (Forrester et al., 1974; Cook et al.,
1966).


43
turkey-baited Bennett traps at both sites. The traps were
operated on 115 different evenings for a total of 211 hours
(Table 3). Cu 1 ico i de s eden i (48.0%) and hinmani
(26.2%) were the most conrnon species and, together, made up
74.2% of the combined catch from each site. Specimens
of s can 1 oni (7.9%), arbor ico1 a (7.6%) and nanus
(5.1%) composed 20.6% of the total. Representatives of the
remaining 7 species made up 5.1% of the total Bennett trap
catch.
Fisheating Creek. During 28 nights of trapping
between December, 1982, and November, 1984, 17,857
specimens of Cu 1 icoides belonging to 13 species were
captured in New Jersey light traps at Fisheating Creek
(Table 4). Cu 1 icoi de s i n s i gnis (44.5%) and edeni
(40.1%) were the most corrmon species and composed 85.2% of
the catch. Cu 1 icoi des knowlton i (7.6%) and O stelIifer
(4.2%) were less conrnon and made up 11.8% of the catch.
Representatives of the remaining 9 species were captured
infrequently or in low numbers and composed only 3% of the
total catch.
During the same period, 2,561 engorged individuals of
5 species of Cu 1 icoi des were captured in turkey-baited
Bennett traps operated on 47 different evenings for a total
of 108 hours (Table 5). Approximately 98% of the total
catch was composed of specimens of edeni (79.6%) and


Figure SI. Seventeen-day-old mega 1oschizont from pectoral
muscle of an experimentally infected
turkey. The megaloschizont contains
numerous elongate and branching cytomeres with
budding merozoites. Hematoxylin and eosin.
Bar 10 um.
Figure 52. Seventeen-day-old megaloschizont from pectoral
muscle of an experimentally infected
turkey. The thick, hyaline outer wall of
the megaloschizont (large arrow) has ruptured,
liberating the merozoites. The merozoites are
spherical and contain a large vacuole (small
arrow). Hematoxylin and eosin. Bar = 10 um.


158
Table 12. Average adjusted measurements of
macrogametocytes. Average measurements of each
variable are expressed as a percentage of the
average uninfected host cell area for that
spec ies .
Var ¡able
Chuchar
Pheasant
Turkey
Parasite Length
43.2
(4.29)+
39.9
(1.59)
43.7
(2.47)
Parasite Width
4.4
(1.11)
3.6
(0.82)
4.9
(0.64)
Paras ite Area
97.9
(10.8)
104.3
(7.71)
117.2
(9.49)
Nucleus Length
8.8
(1.37)
7.9
(1.08)
7.2
(1.29)
Nucleus Width
4.1
(0.83)
3.3
(0.68)
4. 1
(0.59)
Nucleus Area
13.9
(2.36)
1 1 .4
(2.22)
13.3
(2.79)
P i gtne n t *
46.1
(14.4)
53.0
(5.55)
46.9
(6.31)
N =
15
7
15
+ Standard deviation
* Adjusted average of number of pigment granules.


41
buffer with 4% sucrose, pH 7.2. After they were cut into
small blocks, processing continued as above.
Pectoral muscle containing mega 1oschizonts was diced
in primary fixative and processed as described above.


Figure 20. Departures from normal for total monthly
precipitation during 1982, 1983 and 1984 at
Fisheating Creek. Normals are based on
averages of monthly totals between 1951 and
1980 (Climatological Data: Florida, 1982, 1983,
1984).


w


46
end and measured 1.0-2.0 urn in length (Mean = 1.69, SD =
0.33) and 0.5-1.0 um in width (Mean = 0.74, SD = 0.19).
Specimens of 4 species of Cu 1 i coid e s, i.e. C.
paraensis, C. nanus, C. s can 1 on i and baue r i were
capable of supporting partial development of me 1eagridis
and had degenerating oocysts on their outer midguts by 4 to
7 days after taking a blood meal (Table 6).
Degenerating oocysts were smaller than mature, 7-day-old
oocysts and contained large refractile granules (Figure
3). No development was observed in 3 specimens of C.
crepuscu laris (Table 6).
Experimental transmission. Salivary gland sporozoites
from specimens of C_^ e d e n i C_^ h i nma n i and C .
arbor i co 1 a infected 12 of 20, 1 of 6 and 1 of 2 domestic
turkeys, respectively, when inoculated IP or IV. One pool
of 5 specimens of knowltoni, ground in Aedes aegypt i
Ringer's, infected a domestic poult. A second pool of
14 specimens of knowltoni was negative, when inoculated
into another poult. Salivary gland sporozoites from a
single specimen of haematopotus did not infect a
domestic poult.
The prepatent period of all successful infections
ranged from 17-18 days, with peak red cell invasion
occurring by 18-19 days. The rate of growth and morphology
of immature and mature gametocytes was consistent with


1 certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi 1osophy.
Donald J. Forrester, Chairman
Professor of Veterinary Medicine
I certify that I have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi 1osophy.
HT i s
Associate
Med icine
ire i ner ,' Cocha i rman
Professor of Veterinary
I certify that 1 have read this study and that in
my opinion it conforms to acceptable standards of scholarly
presentation and is fully adequate, in scope and
quality, as a dissertation for the degree of Doctor of
Phi 1osophy.
Assistant Professor of Entomology
and Nema to logy


286
Herman, C.M., W.C. Reeves, H.E. McClure, E.M. French and
W.M. Hamnon 1954. Studies on avian malaria in vectors
and hosts of encephalitis in Kern County, California.
Am. J. Trop. Med. Hyg. 3: 676-695.
Howells, R.E., and E.E. Davies. 1971. Nuclear division
in the oocyst of Plasmodiurn berghei. Ann. Trop. Med.
Parasitol. 65: 451-459.
Huff, C.G. 1932. Studies on Haemoproteus of mourning
doves. Am. J. Hyg. 16: 618-62T!
Huff, C.G. 1942. Schizogony and gametocyte development
in Leucocytozoon si mo n di, and comparisons with
P1 asmodium and Haemoproteus. J. Infect. Dis. 71:
18-32.
Huff, C.G. 1969. Exoerythrocytic stages of avian and
reptilian malarial parasites. Exo. Parasitol. 24:
383-421.
Humason, G.L. 1979. Animal tissue techniques. W.H.
Freeman and Co., San Francisco. 661 pp.
Janovy, J.,Jr. 1966. Epidemiology of P1 a smodiurn hexamerium
Huff, 1 935 in Meadowlarks and S t a r 1 i n g~i oT the
Cheyenne Bottoms, Barton County, Kansas. J.
Parasitol. 52: 573-578.
Julian, R.J., T.J. Beveridge and D.E. Galt. 1985. Muscovy
duck mortality not caused by Haemoproteus. J.
Wildl. Dis. 21: 335-337.
Julian, R.J., and D.E. Galt. 1980. Mortality in muscovy
ducks (Cajrina mosciuta) caused by Haemoproteus
infection"! J. Wildl. Dis. 16: 39-44.
Kachigan, S.K. 1982. Multivariate Statistical Analysis.
Radius Press, New York. 297 pp.
Kaye, D., J.G. Merselis and E.W. Hook. 1965. Influence of
P1 a smodium b e rgh ei infection on susceptibility to
s a 1mon e11 a infect ion. Proc. Soc. Exp. Biol. Med.
120: 810-813.
Kellog, F.E., A.K. Prestwood, R.R. Gerrish and G.L.
Doster. 1969. Wild turkey ectoparasites collected in
the southeastern United States. J. Med. Ent. 6:
329-330.


290
Nautical Almanac.
D.C.
Nautical Almanac.
D.C.
Nautical Almanac.
D.C.
1982.U.S. Govt. Print. Off., Washington,
1983.U.S. Govt. Print. Off., Washington,
1984.U.S. Govt. Print. Off., Washington,
Nathan, M.B. 1981. A study of the diurnal biting and flight
activity of Cu 1 i coid e s ph1e bo t omu s (Williston)
(Dipt era: CeratopogonidaeJ using tTTree trapping
methods. Bull. Ent. Res. 71: 121-128.
Newberne, J.W. 1957. Studies on the h i stopatho 1ogy of
Leucocvtozoon simondi infections. Am. J. Vet. Res.
1TF"191-199.
Nolan, R.A., A.H. Brush, N. Arnheim and A.C. Wilson.
1975. An inconsistency between protein resemblance and
taxonomic resemblance: immunological comparison of
diverse proteins from gallinaceous birds. Condor 77:
154-159.
Oliver, J.E.
John Wiley
1 973 CI ima t e
and Sons, Inc.,
and man's environment
New York. 517 pp.
Opitz, H.M., H.J. Jakob, E. Wiensenhuetter and V. Vasandra
Devi. 1982. A myopathy associated with protozoan
schizonts in chickens in commercial farms in
peninsular Malaysia. Avian Path. 11: 527-534.
O'Roke, E.C. 1930. The morphology, transmission, and
life history of Haemoproteus 1ophortyx O'Roke, a blood
parasite of th e Califo r ni a val ley quail. U n i v .
Calif. Publ. Zoo 1. 36: 1-50.
Powell, J.A. 1967. Management of the Florida Turkey
and the Eastern Turkey in Georgia and Alabama. In
"The Wild Turkey and its Management". O.H. Hewitt,
ed. The Wildlife Society, Valley Offset, Inc.,
Deposit, New York. pp. 409-451.
Rogge, D. 1 968 Experi mente11e Bee inf1ussung der
Ha emo proteus Parasitaemie beiin Grunfinken
(Ca r d u e1 is c h1 o ris) durch kunstlichen Langtag.
Acta Parasit. Pol. T5: 397-407.
Rudzinska, M.A., and W. Trager. 1968. The fine structure
of trophozoites and gametocytes of P1 a smodium
coatneyi J. Protozool. 1 5: 73-88 .


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
THE EP1ZOOTIOLOGY AND PATHOGENICITY OF
Haemoproteus meleasridis Levine, 1961,
FROM FLORIDA TURKEYS
By
Carter Tait Atkinson
December, 1985
Chairman: Donald J. Forrester
Cochairman: Ellis C. Greiner
Major Department: Veterinary Medicine
Avian species of the genus Haemoproteus are comnon
haemosporidian parasites found in many families of birds.
In spite of their widespread occurrence, little is known of
their vectors, epizootiology and pathogenicity.
Ha emo proteus me 1e a g rid i s occurs in Wild Turkeys
throughout the southeastern United States. Results of this
study showed that members of at least 5 species of
ceratopogonid flies in the genus Cu 1 i coide s, i.e.
Cu 1 ic o id e s e d e n i Cu 1 icoid e s hinma ni, Cu 1 icoide s
arbor ¡cola, Cu 1 ic o id e s k n ow1t o n i and Cu 1 ic oide s
haematopotus, could support development of the
sporogonic stages of the parasite.
Results of a 2-year epizooti o 1ogica 1 study of
Ha ernoproteus me 1 eagr i d i s indicated that Cu 1 i co i des eden i is
xv i i


241
similar lesions. Instead, he found small schizonts in
endothelial cells of lung tissue. It is not clear from his
work, though, whether he examined histological sections of
skeletal, cardiac and gizzard muscle. The absence of
obvious gross lesions and cysts in the turkey with a
17-day-old infection of me 1eagr i d i s suggests that
megaloschizonts may be missed, particularly if their
development is not suspected. Since muscle tissue makes up
a large proportion of the total body mass, small numbers of
scattered cysts may be difficult to find and yet produce
enough gametocytes to infect a large proportion of the red
blood cells.
The elongate, f i rst-generat i on merozoites of H
me 1 e a g r i d i s are similar to the merozoites contained in
small schizonts described by Gardiner et al. (1984). The
development of schizonts in the spleen of 1 turkey
during the pathogenicity experiment indicates that the
exoerythrocyt ic stages found by Gardiner et al. (1984) may
be part of the life cycle of lophortyx. Since
haemoproteids are believed to be specific to host family
(Bennett et al., 1982), Northern 3obwhites may be a
susceptible but aberrant host for lophortyx.
Clearly, additional experimental work with other species of
Haemop roteus is needed to determine the significance of
these exoerythrocytic stages in haemoproteid life cycles.


271
Since most studies o£ oocysts o£ liaemospor i d i an
parasites are not based on serial sections, it is not clear
when final fragmentation of the large polyploid nucleus
occurs. Howells and Davies (1971) suggested that the
large, lobulate nucleus of berghei may break up soon
after cleft formation subdivides the cytoplasm into
sporoblast bodies. Each nucleus may then undergo a
final division as opposing ends migrate into sporozoite
buds. Schrevel et a 1 (1977) felt that budding and
fragmentation of the nucleus occurred at the same time
sporozoites budded from the sporoblast body. They also
suggested that the multiple mitoses that occurred prior to
sporozoite differentiation allowed numerous genetic
units to be positioned in the nuclear cortex for the
direction of sporozoite differentiation.
Studies of Leucocytozoon have indicated that
nuclear division may occur in a manner similar to
PIasmodiurn. Multiple kinetic centers have been observed in
the nuclear envelope of s imond i L_^ dubreu i 1 i and L.
t a w a ki, but the sequence of events leading to final
fragmentation of the nucleus has not been studied in detail
(Desser, 1972c; Wong and Desser, 1976; Desser and Allison,
1979).
Kinetic centers have not been described in H
metchnikovi, nor were they found in this study.
However, it seems likely that haemoproteids may undergo


Figure 69. Circulating macrogametocyte. The gametocyte
has a cytostome (arrow) with an associated food
vacuole (Fv) derived from an indentation of the
2 outer layers of the pellicle (arrow). Older
food vacuoles (double arrow) are bound by a
single unit membrane and contain diffuse,
electron dense masses of pigment (?). X 37,000.


MONTHS
INCHES INCHES INCHES


23
1 night during a sentinel period, the geometric mean of
all samples was calculated (Bid1ingmeyer, 1969).
A standard New Jersey light trap (Hausherr's Machine
Works, Old Freehold Rd. Toms River, NJ) equipped with
a 40 watt incandescent bulb, an automatic timer and a
delivery cone made from 40-mesh brass was operated at
each of the 2 sites 1 night approximately every 2 weeks.
Two to 3 kg of dry ice were placed in an insulated paper
envelope and hung next to the top of the trap to act as
a carbon dioxide attractant. The trap was operated in
the middle levels of the canopy at the same spot where
the Bennett trap had been placed. Sampling with the New
Jersey trap usually followed or preceded operation of
the Bennett trap by 1 day. The trap was powered by a
portable 500 watt gasoline generator and was started 30
minutes to 1 hour before sunset and run until dawn.
Insects were collected into a 1 pint mason jar containing
10% buffered formalin and a small amount of detergent
(Alconox, Fisher Scientific). Aliquots of each sample
were poured into a white enamel pan and diluted with
water. Specimens of Cu 1 icoides were picked from each
aliquot with a Pasteur pipette, identified, grouped by
sex and parity and counted. New Jersey light trap data
for each species were expressed as the log]g of the number
captured per trap night, plus 1. The geometric mean was


244
post-infection (Figure 37). The surviving high dose birds
remained significantly smaller than control and low dose
turkeys throughout the course of the study (Figure 38).
Low dose birds were smaller than controls, but not
significantly so. If the sample size had been larger,
it is possible that significant effects on growth and
weight gain would have been detected for the low dose group
as well.
Few host effects were associated with the development
of the erythrocytic gametocytes, although these may have
been masked by the massive host response to the
mega 1oschizonts. Average hematocrit and hemoglobin values
were not significantly different for any of the 3
experimental groups at the crisis, or at the second peak in
parasitemia at 6 weeks post-infection (Figures 39, 41). A
significant drop in average hemoglobin concentration
occurred in the high dose group, 1 week after the crisis.
This drop corresponded to the rapid clearance of
parasitized red blood cells from the circulation and their
replacement with immature erythroblasts that had not
completely synthesized their total hemoglobin content
(Lucas and Jamroz, 1961). The absence of other significant
weekly differences in average hemoglobin concentration and
average hematocrit among the 3 experimental groups
indicates that removal of parasitized red cells was
balanced by the synthesis and release of erythroblasts.


12
will need to be synonymized once their life cycles are
better known. Few of the experiments needed to confirm
the current classification have been done. Huff (1932),
Coatney (1933) and Baker (1957, 1966, 1968) studied the
host specificity of sacharovi and 1C macea 11 umi from
Mourning Doves, Zenaida macroura carol inensis, H. co1umbae
from Rock Doves, Co 1 umba 1 i v i a, and pal umbi s from Wood
Pigeons, Co 1umba pa 1umbis, respectively. Huff ( 1 932)
was able to transmit H^ s acha r ov i and H^ macea 11 umi to
Rock Doves by the bite of infected hippoboscid flies,
but Coatney (1933) was unable to infect Mourning Doves
with H^ columbae by either fly bite or inoculation of
sporozoites. Baker (1966b, 1968) attempted unsuccessfully
to transmit fK pa 1 umb i s from Wood Pigeons to Rock Doves
by injection of sporozoites. Similar host restriction
was demonstrated by Fa 1 1 is and Bennett (1960) for H.
canachites f r om a Spruce Grouse, Canachi tes canadens i s.
They inoculated sporozoites obtained from Cu 1 icoides
sphagnume n sis into uninfected Ruffed Grouse, Bonasa
umb e11u s domestic ducks, Anas bo s cha s a Rock Dove,
Columba 1 i via, and a Java Sparrow, Padda oryzivora. Only
Ruffed Grouse became infected. These results must be
interpreted with care since the authors did not include
positive controls when they inoculated the Rock Dove and
Java Sparrow.


240
morphology to the mega 1 osch¡zonts of Leucocytozoon
(Gardiner et al., 1984). In most cases, development of
megaloschizonts occurred in the absence of circulating
gametocytes. Garnham (1973a, 1973b) suggested that
these were aberrant Leucocytozoon infections in abnormal
hosts. Gardiner et al. (1984) recently reviewed these
reports and described similar mega 1oschizonts in pen-reared
Northern Uobwhites from Ca1ifornia. The cysts described in
all the reports have a number of features in common,
including a hyaline wall of variable thickness, development
in muscle tissue and formation of numerous, spherical
merozoites approximately 1 um in diameter. Most have an
associated host myopathy.
The results of this study provide the first
experimental evidence that some of these organisms may
be haemoproteids Gardiner et al. (1984) observed
pigmented gametocytes, morphologically similar to H.
1ophor t y x, in red blood cells of Northern Bobwhites that
died from an extensive myositis associated with developing
mega 1 oschizonts However, since they observed small
schizonts with elongate zoites and hypertrophied host cell
nuclei in spleen tissue that were unlike any reported
schizonts of Leucocytozoon or Haemoproteus, they suggested
that the causitive organism may be a new member of the
Haemosporina. O'Roke (1930) conducted a detailed study of
H. 1ophortyx in California Valley Quail, but never reported


26
by site, year and quarter. To determine whether capture
data from each site, year and quarter could be combined
by species and analyzed as a single data set, the mean
capture time, standard deviation and sample size was
computed for each species by site, year and quarter.
Because some species were not active during all quarters,
and a number of empty "cells" were present when mean
capture times were examined by site, year and quarter,
it was assumed that year had a minimal effect on
variability in the data. Data for each species from each
of the 3 years of the study were combined by site and
quarter. An analysis of variance using the Statistical
Analysis System, general linear models procedure was used
to test for significant interaction effects between
species*si te, spec i es*quir ter, s i te*quar ter and
spec ies*site*quarter (SAS User's Guide: Statistics, 1982).
An alpha level of 0.05 was considered significant.
Species*site and species*quarter interactions were
significant (p< 0.0001). The analysis was repeated without
the Fisheating Creek data to determine whether data from
the 2 sites at Paynes Prairie could be combined by
quarter. Species*quarter interactions were significant
(p < 0.0001). The final analysis was done by site (Site
A and B combined) and quarter with a one-way analysis
of variance using the SAS general linear models procedure
(SAS User's Guide: Statistics, 1982). Significant species


Figure 38. Average tarsometatarsal lengths of high dose birds
(Q"""0 ), low dose birds ( A A ) and control birds
(@ ). Statistically significant differences among
groups for each week of the study are indicated by
the boxed points (p 0.05).


Figure 70. Maturing macrogamete. The outer layer of
the pellicle has separated from the parasite
and formed membranous coils (arrows).
Mitochondria (M) and endoplasmic reticulum
filled with amorphous material are scattered in
the cytoplasm. X 58,000.


Page
RESULTS 42
Epizoot¡ology 42
Vectors 42
Paynes Prairie 42
Fisheating Creek 43
Experimental Infections 44
Sporogonic development 44
Experimental transmission 46
Activity Cycles 47
Bennett trap catches 47
New Jersey trap catches 49
Sentinel Study 50
Paynes Prairie 50
Fisheating Creek 51
Transmission and Vector Abundance 52
Paynes Prairie 52
Fisheating Creek 54
Estimation of Prevalence 55
Pathogenicity 96
Experiment 1 Pathology 96
Gross observations spontaneous deaths 96
Gross observations surviving birds ... 97
Microscopic observations -
spontaneous deaths 100
Microscopic observations -
surviving birds 103
Paras i temi a 104
Weight 105
Tarsometatarsal length 106
Hematocr it 106
Plasma protein concentration 107
Hemoglobin 108
Experiment 2-Exoerythrocytic Development .. 109
Three days 109
Five days 109
Eight days 110
Eleven days 1 1 1
Fourteen days 112
Seventeen days 112
Exoerythrocytic Development Natural
Infections 114
Hos t Spec ificity 151
Parasitemia 151
v i


110
adjacent muscle fibers. The fibers were swollen, pale and
hyaline and often had disrupted sarcoplasm (Figure 54).
Individual swollen, rounded and hyaline muscle fibers were
scattered randomly throughout healthy tissue. A few focal
areas of perivascular monocytic infiltrate were present.
Schizonts measuring from 12 to 20 um in diameter were
present both within and between muscle fibers (Figures 43,
44). The smaller, more immature parasites contained
dark-staining, granular cytoplasm (Figure 43). The larger,
more mature forms were packed with dark-staining, elongate
zoites that were bent and twisted around each other (Figure
44). None of the parasites had an associated host
reaction.
Sections of heart from the infected bird had focal
aggregates of mononuclear cells scattered randomly
throughout the tissue. Hepatocellular atrophy and necrosis
were evident in sections of liver. The spleen was enlarged
and contained numerous erythropoietic cells in the vascular
sinuses. Other tissues from the infected and control bird
were unremarkable.
Eight days. By 8 days post-infection infected birds
showed dramatic signs of improvement and were back on their
feet. Sections of skeletal muscle from 1 infected bird had
numerous focal areas of necrotic muscle fibers, infiltrated


LOG|Q (N+ I) LOG ¡o IN II
oo
OJ


109
weeks PI, among 0, 3 and 5 weeks PI and among I, 2 and 5
weeks PI.
Average hemoglobin values within the control group
were significantly lower at 6, 7 and 8 weeks PI than
they were at week 0. Average values at week 0 were
significantly lower than those at 3, 4 and 5 weeks PI.
Average values at 3, 4 and 5 weeks PI were significantly
lower than those at 1 and 2 weeks PI.
Experiment 2 Exoerythrocyt i c Development
Three days. At 3 days post-infection, sections of
skeletal muscle from the inoculated poult had a few
focal areas of perivascular monocytic infiltrate. Some
muscle fibers appeared disrupted, with granular sarcoplasm
and scattered islands of pale, disorganized myoglobin
(Figure 53). A single, uninucleate parasite with blue-gray
granular cytoplasm was detected within a capillary (Figure
42). The organism was 3 um in diameter and present in only
1 of a series of 4 um serial sections. Other tissues in
the infected and control bird were unremarkable.
Five days. By 5 days post-infect ion, infected poults
developed severe lameness in both legs. All had difficulty
standing and moving about their cage, while control
birds remained active. Skeletal muscle from an infected
bird had focal areas of necrosis that involved many


198


44
h i runa ni (18.6%). The remaining catch was made up of
specimens of knowltoni (1.4%), arbor ico1 a (0.5%) and
C. bauer i (0.04%).
Experimental Infections
Sporogonic development. Engorged individuals of
10 species of Cu 1 icoi des were collected from Bennett traps
baited with turkeys infected with 1G me 1eagr i d i s Fresh
preparations and Giemsa-stained smears of engorged midguts,
dissected from specimens of edeni within 24 hours after
a blood meal was taken, had numerous ookinetes. In
fresh preparations, ookinetes had a retractile "knob" or
"point" at one end and a mass of golden-brown pigment at
the other (Figure 1). Fifteen Giemsa-stained ookinetes
measured 16.5-21.1 um in length (Mean = 18.95, SD = 1.4)
and 2.5-3.75 um in width (Mean = 2.98, SD = 0.458). The
nucleus was oval to round and measured 2.0-3.25 um in
length (Mean = 2.48, SD r 0.417) and 1.5-2.25 um in
width (Mean = 1.98, SD = 0.32). In well stained
preparations, 1, and occasionally 2, empty vacuoles,
slightly smaller than the ookinete nucleus, were located
anterior and/or posterior to it.
Dissections of specimens of Cu 1 icoi des from 2 to 7
days after they had taken blood meals from infected
turkeys, revealed both viable and degenerating oocysts


-101 2345678
WEEKS
Otl


Figure 53. Disrupted muscle fiber (arrow) from pectoral
muscle of a turkey with a 3-day-old
experimental infection of Ha emo proteus
me 1eagridis. Hematoxyl in and eosin. Bar =
10 um.
Figure 54. Swollen, hyaline and disrupted pectoral muscle
fibers from a turkey with a 5-day-old
experimental infection of Ha emo proteus
me 1 ea^r i d i s Normal muscle fibers (M)sur round
the lesion. Hematoxylin and eosin. Bar =
10 um.
Figure 55. Regenerating muscle fibers (arrows) from
pectoral muscle of a turkey with an
8-day-old experimental infection of
Ha emo proteus me 1eag rid i s Hematoxylin and
eosin.3ar = 10 um.
Figure 56. Deteriorating 17-day-old megalosch i zont (S)
from pectoral muscle of an experimentally
infected turkey. The mega 1 oschizont (S) is
surrounded by giant cells (G) and an outer
layer of connective tissue (arrow).
Hematoxylin and eosin. Bar = 100 um.


22
mesh (28 mesh per cm). Two domestic turkeys that were
the same age and size were hoisted alternately into the
canopy. While 1 bird was exposed, the second was left
undisturbed under the screened wooden frame for 10 minutes
to allow specimens of Cu 1 icoi des to complete their blood
meals. Engorged and unengorged individuals of each species
of Cu 1 icoides were then aspirated through a sleeve in
the top of the outer screened frame as they rested on
its interior.
Specimens of Cu 1 icoi des aspirated during each run
of the Bennett trap were placed into half-pint cardboard
cartons with screened tops and supplied with a cotton
pad moistened with 5% (w/v) sucrose. The beginning and
ending times for each run of the Bennett trap, temperature,
wind velocity estimated by the Beaufort scale (Oliver,
1973) and overall weather conditions were recorded. Since
early attempts at Bennett trapping were unsuccessful in
rain and when wind velocity exceeded Beaufort 3 (4-7
miles/hr.), all trapping throughout the course of the
study was done on calm evenings when rainfall was not
imminent. Between 1 and 9 days after capture, the
specimens of Cu 1 icoi des were identified (Blanton and Wirth,
1979) and classified according to parity by the method
of Dyce (1969). Bennett trap catches of individuals of
each species were expressed as the 1 og ] q of the number
captured, plus 1. When sampling was done on more than


LOGlo (N + I) LOGlo (Nt I) LOG10(Ntl)
79
17 19 21 23 01 03 05 07 09 II 13 15
HOUR


217
turkeys, exposed in Wild Turkey habitat. It is similar to
the shorter prepatent periods of other haemoprote i ds
transmitted by species of Cu 1 i co i des ; 14 21 days for
H. ne11ionis, 14 days for ma n s o ni and 11-14 days for H.
ve 1 an s (Fall is and Wood, 1957; Fall is and Bennett, 1960;
Khan and Falls, 1971).
Activity Cycles
Most species of Cu 1 i coide s have crepuscular and/or
nocturnal peaks in their daily activity cycles that may
play an important role in their contact with potential
hosts (Kettle, 1965, 1977; Barnard and Jones, 1980).
Kettle (1968b) noted that this cycle is probably an
endogenous circadian rhythm, regulated by changes in light
intensity and modified by local meteorological
conditions such as wind velocity and temperature. A number
of sampling techniques have been used to measure diel
changes in activity, including truck-mounted
interception traps (3 id1ingmeyer, 1961), suction traps
(Service, 1971), biting collections (Kettle, 1968a) and
paddle traps (Nathan, 1981). Studies that have employed
more than 1 sampling method have generally found them to be
in close agreement in detecting major peaks of activity
during the 24-hour cycle (Nathan, 1981; Service, 1971).


98
muscles. They were 2-3 times larger and were more diffuse
than cysts observed in the 4 fatal infections. Cysts were
not detected in any of the 11 control birds.
All surviving birds had focal areas of reddened,
swollen mucosa and occasional petechial hemorrhages
throughout the length of the gut. None of the birds
exhibited clinical signs of salmonellosis, i.e. "pasty"
vent, bloody diarrhea or depression. The "pasty vents"
observed in high dose birds at 1 week PI resolved
spontaneously in surviving birds by 4 weeks PI. At the
termination of the experiment, cloacal swabs from 2 of
11 control birds, 1 of 12 low dose birds and 1 of 8 high
dose birds were positive for Salmone11 a enteriditis
Group 8. Coccidian oocysts were not detected in pooled
fecal samples at either 4 or 8 weeks post-infection.
Incidental findings included the presence of numerous
white nodules, approximately 5 ¡ira in diameter, that were
scattered throughout the mesentery lining the abdominal
cavity of 1 control bird and 2 high dose birds.
Average wet weights of hearts and livers removed from
the birds at necropsy were not significantly different for
any group (Table 11). The difference in average spleen
weights among the 3 groups was highly significant (p<
.0002).


Figure 37. Average weights of high dose birds ( O O ), low
dose birds ( A--*-A ) and control birds ( 9 9 ).
Statistically significant differences among groups
for each week of the study are indicated by the boxed
points (p 0.05).


237
is associated with development of the large mega 1oschizonts
(Miller et a 1 ., 1983; Newberne, 1957).
During the earliest stages of infection with H
me 1eagr i d i s myopathy was restricted to isolated muscle
fibers. Sy 5 days post-infection, entire bundles of as
many as 5-10 adjacent myofibers were necrotic. The absence
of focal areas of inflammation and necrosis irrmediately
around f i rst-generat ion schizonts suggests that the
extensive necrosis was not due to the release of toxins by
developing parasites or to blockage or interference with
the local circulation. The extensive myopathy may have
been related to the size of the initial inoculum. The
total numbers of exoerythrocytic schizonts were
significantly fewer than in the pathogenicity experiment
where the total sporozoite dose was only 1/3 as large.
Perhaps many host cells were invaded by more than 1
sporozoite and were unable to support the development of
multiple schizonts. Their early death would explain the
low number of megaloschizonts that developed relative to
the infective dose. Wallace (1973) observed myositis
and myocarditis in mice that had been infected orally with
large numbers of Sarcocys t i s sporocysts. He found that the
myositis became apparent before sarcocysts developed and
was not associated with the early first-generation
schizonts. The myositis was also evident in experimentally
infected mice that failed to develop detectable sarcocysts,


LIST OF TABLES
Tab Ie Page
1 Proven and presumed vectors of avian
haemoproteids 4
2 New Jersey light trap collections Paynes
Prairie 57
3 Engorged specimens of Cu 1 icoi des captured in
Bennett traps at Paynes Prairie 58
4 New Jersey light trap collections Fisheating
Creek 59
5 Engorged specimens of Cu 1 i coi des captured in
Bennett traps at Fisheating Creek 60
6 Susceptibility of wild-caught specimens of
Cu 1 icoi des to Haemoproteus me 1eagridis 61
7 Mean capture times for specimens of Cu 1 icoi des
taken in Bennett traps at Paynes Prairie and
Fisheating Creek 62
8 Yearly prevalence of Haemoproteus me 1eagrid i s
in specimens of Cu 1 icoi des edeni at Paynes
Prairie 63
9 Yearly prevalence of Haemoproteus meleagridis
in specimens of Cu 1 icoi des edeni at Fisneating
Creek 64
10 Attempted isolations of Haemoproteus
me I eagrid i s from pools of Culi coi des hinmani,
Culicoides arboricola and Cu 1 icoi des
knowl ton i 65
11 Average organ weights at necropsy 99
v i i i


Figure 59. Parasitemias per 10,000 red blood cells for turkeys,
(O O ), the Chuckar ( A A ) and the Ring-necked
Pheasant ( @ ) with experimental infections of
Ha einop roteus me 1 e a g r i d i s Values for turkey were
calculated as the average parasitemia for the 2 infected
birds. All birds received the s ame n umb e r of
sporozoites.


ACKNOWLEDGEMENTS
I would like to express my sincere appreciation to all
the people who made this study possible. Donald J.
Forrester and Ellis C. Greiner conceived the initial
idea and recruited me as a graduate student. Lovett E.
Williams, Jr., and David Austin provided advice and
assistance in selecting the study sites at Paynes
Prairie State Preserve and at Fisheating Creek. Logistical
support from David Austin and the Florida Game and Fresh
Water Fish Commission as well as cooperation from Ben
Swendsen and Lykes Brothers Corporation made the field work
at Fisheating Creek possible. James Richardson and
Claus Buergeldt provided invaluable help with initial
necropsies and the interpretation and description of
microscopic lesions. Randolf Carter, Laura Perkins and
Debbie Schons contributed advice on the statistical
analysis. The IFAS Central Electron Microscopy Facility
provided materials and equipment for studies of the fine
structure of Ha emo proteus me Ieagridis.
Jerome Dj am deserves thanks for the hundreds of
turkeys he helped to bleed, for his innumerable trips to
Paynes Prairie to feed sentinel birds and for his patience


Figure 83. Three-day-o1d oocyst {coin a specimen of
Cu 1 icoi des edeni The oocyst is surrounded by
a thick, amorphous wall (large arrows) and is
located beneath the basement membrane (3m) of
the midgut. The pellicle of the parasite
is underlaid by osmiophilic thickenings (small
arrows). Nuclei (N) with prominent nucleoli,
large membrane-bound, lipid-like inclusions
(L) and mitochondria (M) with tubular cristae
are present. X 26,100.


272
nuclear divisions in a manner similar to ?1 a smodi um and
Leucocytozoon.
Crystal 1 oid Crystalloid aggregations of electron
dense particles consisting of a lipo-protein complex
have been reported from the oocysts of Leucocytozoon, H.
me t c h n i!: o v i and from the early oocysts of EC ga 1 1 i naceuin
and EC b e r g h e i (Trefiak and Desser, 1973). Trefiak and
Desser (1973) found that they they originate from amorphous
aggregations of electron dense material in the macrogametes
o f L_^ s i mond i They suggested that the crystalloid
functions as an energy source since it appears to be
utilized quickly in rapidly growing oocysts of
P1 a smodium and is incorporated into the sporozoites of
Leucocy tozoon and me t chnikovi. The former have been
found to persist in their avian hosts for up to 11 days and
may require a reserve energy source to maintain their
viability (Khan et al., 1969).
Membrane-bound lipid inclusions have been reported in
the oocysts of !C cynomo1gi, EC ga11inaceum, EC fa 1 c i parum,
L s i mo n d i dub r eu i 1 i and 1^ tawak i (Terzakis, 1971;
Terzakis et al., 1966; Sinden and Strong, 1978; Desser,
1972c; Wong and Desser, 1976; Desser and Allison,
1979). They were not reported in oocysts of 1C metchnikovi
(Sterling and DeGiusti, 1974). Generally, lipid inclusions


222
limited, though, by the availability of a suitable domestic
host for the blood parasite under investigation. As a
result, this technique has been limited primarily to the
study of Leucocytozoon and Haemoproteus in ducks.
Forrester (unpublished data) and Akey (1981), at
Fisheating Creek, were the first to use domestic, sentinel
turkeys to monitor the natural transmission of H
me 1eagridis. Earlier surveys in this area by Forrester et
al. (1974) established that me 1eagr i d i s had a prevalence
that ranged from 70 100% in Wild Turkeys older than 5
months. The increasing prevalence of me 1 eagrid i s in
juvenile birds collected at monthly intervals up to 1 year
after hatching suggested that transmission of the parasite
occurred throughout the year (Forrester et al., 1974). In
concurrence with this, Forrester (unpublished) found
that transmission occurred to 50- 100% of the sentinel
poults that he exposed for 12, 2-week intervals between
May, 1978 and May 1979. Akey (1981) had similar results
during a shorter study between May and August, 1980, in the
same area. Both Forrester (unpublished) and Akey
(1981) noted that transmission of me 1eag rid i s
occurred at higher prevalences among birds caged in the
canopy, than to those caged on the ground. Results of the
present study confirm these previous observations at
Fisheating Creek and extend the observations from the warm,


224
or whether a significant reduction in the population
size had occurred.
Fisheating Creek is subject to frequent flooding
and wide fluctuations in water levels and flow (U.S.
Geological Survey, Water Resources Division, Orlando,
Florida). During the drier months of the year (November -
March), flow may be reduced to a trickle and the stream can
become a series of shallow pools and ponds. During
heavy rainstorms, stream levels may rise as much as 7 feet
in 24 hours. Because the topography in the area is very
flat, ranging from 30 to 55 feet above sea level, the
creek swamp may be flooded up to 1 km or more on either
side of the stream channel. Since individuals of many
species of Cu I i c o i d e s including C^ eden i breed in
moist soil or mud at the edges of permanent or
semi-permanent bodies of water (Kettle, 1977),
extensive flooding may erode breeding sites and wash the
immature stages away. Corresponding reductions in adult
populations may reduce the numbers of potential vectors to
below the critical density needed to maintain continuous
transmission of the parasite.
Patterns of transmission and vector abundance at
Paynes Prairie were similar to those at Fisheating Creek.
The year-round activity of individuals of Ch edeni, the
isolation of me 1ea3ridis from specimens of naturally
infected O e d e ni and the conspicuous absence of


273
tend to be smaller in oocysts of those species that form
prominent crystalloids.
The absence of a crystalloid in 3-day-old and
6-day-old oocysts of me 1eagridis suggests that the large
lipid-like inclusions observed in this study may
function as energy reserves. Observations of earlier
oocysts are needed to determine whether a crystalloid is
present in this parasite.
Mega 1oschizon t s
The thick laminated cyst wall and the process of
inerozoi te formation within mature mega 1 osch i zonts of H,
me 1eagr i d i s appears to be unique among haemosporidian
parasites. Observations of me ga 1 oschizonts of L.
s i mo nd i and the meg a 1 o s c h i zo n t s of the unidentified
haemosporidian from Northern Bobwhites are most similar to
those of this study. Desser (1970a) reported a capsule
around mega losch i zon t s of s i mo n d i composed of a meshwork
of reticular fibers and an outer fibrous layer. The
capsule was external to the host cell plasma membrane.
Desser and Falls (1967) suggested that it was primarily of
host origin and probably secreted by fibroblasts that were
often found surrounding the mega Ioschizonts and their host
cells. Gardiner et a 1 (1984) found that
mega 1oschizonts of the parasite of Northern Bobwhites were
surrounded by a moderately dense, amorphous wall that


165
Table 19. Classification sumnary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells infected
with macrogametocytes Discriminant scores were
classified with a discriminant function derived
from the calibration data set summarized in
Table 18.
Classified Into Species
Spec ies
Ch u c k a r
Pheasant
Turkey
Other
Tota 1
Chucka r
11 (73.3)+
0 (0.0)
3 (20.0)
1
(6.7)
15
(100)
Pheasant
3 (42.9)
0 (0.0)
3 (42.9)
1
(14.3)
7
(100)
Turkey
3 (20.0)
0 (0.0)
10 (66.7)
2
(13.3)
15
(100)
Total
17 (46.0)
0 (0.0)
16 (43.2)
4
(10.8)
37
(100)
Priors*
0.4054
0.1S92
0.4054
+ Standard Deviation
* Prior probability of being assigned to that class


218
The large, evening Bennett trap catches of
specimens of eden i CC hinmani, C. arboricol a and C.
knowlton i suggest that individuals of these species are
primarily crepuscular with major peaks in biting
activity near sunset. Since sampling with the Bennett
traps was restricted primarily to the 2 hours preceding and
following sunset, peaks of activity at other times of
the day would have been missed. However, the limited
amount of Bennett trapping that was done at night, at dawn
and during the day did not detect any major periods of
activity.
Peak evening collections of specimens of O hinmani
were consistently and significantly earlier than
collections of the other species (p < 0.05). Differences
among mean capture times for specimens of eden i C.
arboricol a and knowltoni were not consistent from
quarter to quarter or site to site. The large amount of
overlap among capture times for individuals of these
species and the correspondingly large standard deviations
for mean capture times at each quarter and site, suggests
that the differences are not biologically significant.
The Bennett trap catches and the CC>2_baited suction
trap catches of specimens of edeni and hinmani
were similar at Fisheating Creek. Individuals of both
species had peaks in their biting/host-seeking activity in
the forest canopy during the 2-hour sampling period that


152
occurred by 20 DPI. Gametocytes were cleared from the
circulation by 36 DPI. The Ring-necked Pheasant that
developed a patent infection proved to be the least
susceptible to me 1eag r i d i s The parasitemia reached
a peak of only 22 parasites per 10,000 red blood cells
at 20 DPI and was cleared from the peripheral
circulation by 26 DPI (Figure 59).
The 2 turkeys inoculated in the second series of
experimental infections were the only birds that developed
patent infections of me 1 e a g r i d i s The Northern
Bobwhites, chickens and all negative control birds remained
negative throughout the course of the experiment.
Morphometric Analysis
Macrogametocytes. Macrogametocytes from each of
the infected host species were morphologically similar
(Figures 60, 62, 64). All completely encircled the host
cell nucleus and were consistent with descriptions of
neotypes of me 1eagridis (Greiner and Forrester, 1980).
The average adjusted cell area of macrogametocytes from the
turkey was larger than that of macrogametocytes from either
the Chuckar or the Ring-necked Pheasant (Table 12).
Average values of other adjusted variables were within 1
standard deviation of each other (Table 12).


19
(2) study the pathogenicity o£ me 1eagridis in
domestic turkeys with controlled experimental infections,
(3) examine the host specificity of the parasite
and
(4) study the fine stucture of the me 1 eag ridi s
in the vertebrate and invertebrate hosts.


102
and 20 DPI lacked similar deposits. Spleen sections
from all 4 birds had extensive areas of follicular
hypoplasia caused by a large reduction in the size of
the periarterial lymphatic sheaths. Numerous mature and
ruptured schizonts were present in reticular cells in
the spleen of 1 bird. These schizonts were smaller in
diameter and lacked the thick, hyaline wall that surrounded
megaloschizonts from muscle tissue (Figures 31, 32). Most
averaged from 10 to 15 urn in diameter and contained small,
spherical zoites, less than 1 urn in diameter. A few
contained elongate zoites.
Heart sections from 1 bird had several thrombi with
associated giant cells and a single, large mega 1oschizont.
Incidental findings in lung tissue from 3 of the 4
birds included the presence of large granulomas composed of
giant cells, mononuclear cells and heterophils that
surrounded large, amorphous eosinophilic central cores.
Fungal hyphae resembling As pe r gi 11u s sp. were evident
in sections from 1 bird. Epithelial cells lining alveolar
capillaries were hypertrophic and associated capillaries
were congested. A fibrino-hemorrhagic exudate containing
macrophages and heterophils flooded alveolar spaces. Some
large blood vessels were occluded by thrombi. Sections of
air sacs of 1 bird and mesentery of another had similar
granulomas that surrounded eosinophilic masses
containing fungal hyphae.


219
included sunset. Suction trap catches of specimens of
C. e d e n i and hinmani fell rapidly in numbers
following twilight and remained low until after sunrise,
when catches of individual hinmani had a second, smaller
increase in numbers. Individuals of both species were
active at lower levels during the day on 1 or more of the 3
collection dates .
Few specimens of eden i and no specimens of C.
h'nmani were captured in the CC>2-baited suction traps
operated near the ground. However, the numbers of
individual eden i observed on the exposed sentinel
turkeys during the day, as well as the high prevalence
of transmission of me 1 e a g r i d i s to sentinel birds
caged on the ground, indicates that ground-level biting
activity occurred. The ground traps may have been located
in a poor position or the technique may not have been
sensitive enough to detect low numbers of Cu 1 icoi des. Snow
(1955) noted that biting activity of individuals of C.
paraensis, C. spinosus and borinqueni spread upward
along the main trunk of the tree and then outward into the
canopy during the day. Since the ground traps were not
positioned near tree trunks, they would not have
detected similar movements by specimens of eden i or
C. hinmani Other studies have found distinct differences
in the vertical distribution of Cu 1 icoides species in
forest habitat (Snow, 1955; Bennett, I960; Service,


90
80
70
60
50
40
90
80
70
60
50
40
90
80
70
60
50
40
89
1982
i iii ii i I I I l 1
J FMAMJ JA S 0 N D
1983
i i i i i i I I i l I l
J FMAMJ J ASOND
1984
_i i i i i i i i 1 1 1 1
JFMAMJJ ASOND
MONTH


Figure Page
69 Circulating macrogametocyte 186
70 Maturing macrogamete 188
71 Exf1 age 11 ating microgametocyte 190
72 Higher magnification of Figure 71 190
73 Maturing macrogamete 192
74 Maturing macrogamete 192
75 Extracellular maturing macrogamete 194
76 A higher magnification of a portion of
Figure 75 194
77 Extracellular exf1 age 11 at ing mi ergametocyte .. 196
78 Extracellular exf1 age 11 ating mi ergametocyte .. 198
79 Extracellular exf1 age 11 ating microgametocyte .. 198
80 Extracellular exf1 age 11 ating microgametocyte .. 200
81 Extracellular exf1 age 11 ating microgametocyte .. 200
82 Cross sections of microgametes 200
83 Three-day-old oocyst from a specimen of
Cu 1icoi des edeni 202
84 Six-day-old oocyst from a specimen of
Culicoi des edeni 204
85 Six-day-old oocyst from a specimen of
Culicoi des eden i 204
86 Six-day-old oocyst from a specimen of
Cul i co i des eden i 206
87 Six-day-old oocyst from a specimen of
Culicoi des edeni 208
88 Megaloschizont and associated phagocytic
cells from pectoral muscle of a high dose
bird that died spontaneously at 20 days
post-infection 210
xv


178
thickenings beneath the pellicle (Figure 84). As the buds
became more elongate, apical organelles including polar
rings, electron dense rhoptries and subpe1 1 i cu 1 ar
microtubules differentiated from the sporoblast body
(Figure 8S). One budding sporozoite had at least 18
subpe 11 i cu 1 ar microtubules in cross section (Figure
85). The pellicle of developing sporozoites consisted
of an outer unit membrane that was underlaid by 2 unit
membranes in close apposition to one another (Figure 85).
Large nuclei were located at the periphery of the
sporoblast body, underneath the budding sporozoites.
Remnants of the apical complex of the ookinete, consisting
of an electron dense canopy and supporting microtubules
were present within the sporoblast (Figure 84). Large
lipid-like vesicles were clustered near the center of
the sporoblast body (Figure 84).
As sporozoites completed budding, the sporoblast body
became progressively smaller, leaving a residual mass
composed of large, lipid-like vesicles and amorphous
cytoplasm (Figure 86). At maturity, oocysts ruptured,
liberating sporozoites into the haemocoel of the
insect. Ruptured oocysts contained the degenerating
remnants of the residual body (Figure 87).


55
with several peaks between April and November of each
year. Culico ides arboricola was captured rarely in Bennett
traps and had only 1 major peak in biting activity in May,
1984 (Figure 18).
Culicoide s knowlton i was primarily a late-spring,
early-summer species, with major peaks in abundance and
biting activity between April and July, 1983, and May and
June, 1984. Fall peaks in biting activity and abundance
occurred in October, 1984.
Estimation of Prevalence
Between May, 1982, and April, 1983, unengorged
specimens of Culicoi des captured in Bennett traps at Paynes
Prairie were dissected and examined for sporozoites in the
salivary glands. Seventeen of 17 specimens of C nanus, 9
of 9 C crepuscu1aris, 33 of 33 C^ arbor ico1 a and 12 of 12
C h i n m a ni were negative for salivary gland
sporozoites. Sporozoites were found in a single,
unengorged specimen of C^ edeni, captured on 25 August,
1982, at Site A. Dissections of 209 other unengorged
specimens of C^ edeni, collected between May and April were
negative. Unfortunately, an uninfected recipient turkey
was not available for inoculation of the sporozoites to
confirm their identification.


3
o£ O nubecu1osus but neither the avian host nor the
vector are available to workers in this country.
Currently, vectors have been reported for fewer than
10% of the described species of Haemoproteus (Table 1).
Five of these, i.e. co1umb a e, H. sacharovi, H.
ma c c a 1 1 um i H lophortyx and FL pa 1 umb i s have been
transmitted by the bite of hippoboscid flies. The
remaining 7 species are presumed to be transmitted by
ceratopogonid flies in the genus Cu 1 ico i de s Several
workers have reported complete development of the
sporogonic stages in 8 species of Cu 1 icoi des and have
transmitted the parasite by i ntraperitonea1 or intravenous
inoculation of sporozoites into suitable, uninfected hosts
(Bennett et al., 1965; Kahn and Fa 1 I is, 1971; Miltgen
et al, 1981; Atkinson et al., 1983). Transmission by
bite has not been demonstrated for any haemoproteid in
this group.
The discovery that hippoboscid flies could transmit
Haemoproteus their common occurrence on birds and the
morphological similarities among gametocytes of various
species, led to the early assumption that this group was
fairly homogeneous. The recent advances in understanding
of the complex life cycles of cyst forming coccidian
parasites such as Sarcocystis and Toxop1 a sma, has
demonstrated that sporozoan life cycles are far more


229
under hormonal control and occurs in conjuction with the
reproductive cycles of their avian hosts (Haberkorn, 1968;
Desser et al., 1968; Rogge, 1968; Applegate, 1970).
Transmission of the parasites occurs during a relatively
short period when recently hatched, susceptible young
are leaving their nests, when vector populations have
increased with the onset of warmer weather and when the
older adult population, with chronic, relapsing infections,
is available as a reservoir of infection (Janovy, 1966;
Falls and 3ennett, 1966).
The more temperate climate at Paynes Prairie, the
smaller and more variable vector populations and the lower
prevalence of infection in the host population are similar
to conditions of hyperendemic, unstable malaria
(Wernsdorfer, 1980). Wide variations in the prevalence of
infection may occur, depending on environmental conditions,
host density and vector abundance and bionomics. However,
in contrast to the more unstable, epidemic transmission of
other avian blood parasites in northern North America,
transmission can remain high throughout most of the year.


106
0, 1 and 2 weeks PI, at 1, 2 and 3 weeks PI and at 3 and 4
weeks PI (Figure 37).
Tarsometatarsal length. Statistical analysis of
tarsometatarsal length revealed that all 4 variables in the
model statement were highly significant (p = .0001). When
comparisons were made by week, all 3 groups were
significantly different at the crisis, at 3 weeks PI. The
high dose group had average tarsometatarsal lengths that
were significantly shorter than control and low dose birds
at all other weeks PI. Other differences between the
control and low dose groups were not significant (Figure
38).
When comparisons were made within group, all groups
showed a significant increase in tarsometatarsal length for
each week.
Hema tocrit. Statistical analysis of hematocrit
revealed that all variables in the model statement were
highly significant (p = .0001) except treatment*week (p
= .6886). Comparisons among the 3 experimental groups,
averaged over all weeks, were not significant (p = .1617).
Comparisons among weeks, averaged over all groups,
showed no significant differences among weeks 0, 6 and
8, among weeks 3, 6 and 7, among weeks 1, 4 and 5 and
between weeks 2 and 4 PI (Figure 39).


13
Miltgen et al. (1981) unsuccessfully attempted to
infect a parakeet, Me 1 opsi11acus sp., with hL desseri
from Blossom-Headed Parakeets by intraper i tonea 1
inoculation of sporozoites obtained from nubeculosus
Fine Structure
For the most part, u1trastructura 1 studies of the
genus Haemoproteus have been limited to the mature and
exf 1 age 11 ating gametocytes of columbae Work by
Bradbury and Roberts (1970), Bradbury and Trager (1968a,
1968b), Gallucci (1974a, 1974b) and Sterling and Aikawa
(1973) has demonstrated that the morphology of these stages
is consistent with that of other haemosporidians.
Studies of the ookinetes of co 1 umbae and velans
have shown that they are similar to each other in structure
and organization, but with several important differences.
Gallucci (1974b) hypothesized that the ookinete is a
conservative stage in the life cycle of the parasite,
since it is produced sexually and more likely to retain
the organelles of its primitive ancestor. Hence,
differences between the ookinetes of these 2 species may
lend credence to the idea that haemoproteids transmitted
by Cu 1 i coi des belong in a separate taxonomic group.


28
saline or Aedes aegypti ringers (Hayes, 1953). The flies
were removed from their cartons with an aspirator and
blown into a small petri dish containing saline and a
drop of Triton 100 X (Fisher Scientific). The specimens
of Culico ides were then transferred to a glass microscope
slide and identified by wing pattern under a dissecting
scope (Blanton and Wirth, 1979). The salivary glands
and midgut were carefully removed with fine dissecting
pins (minuten nadeln) in a drop of saline, covered with
a coverslip and examined at 400X for oocysts and
sporozoites with Norma r ski contrast interference
microscopy. Questionable identifications of individual
specimens of Culicoides were confirmed after dissected
flies had been cleared overnight in liquid phenol and
mounted on a microscope slide in a drop of 50% liquid
phenol and 50% Canada balsam. Engorged midguts were
smeared on glass slide, air dried, fixed in absolute
methanol and stained with Giemsa as described earlier.
Measurements of ookinetes were made from camera lucida
drawings and adjusted to scale with a slide micrometer.
Domestic poults were infected by drawing salivary
glands containing sporozoites into the needle of a
tuberculin syringe and injecting them int r ape r i tonea 1 1 y
(IP) or intravenously (IV) into uninfected poults. Whole
flies were also ground in 0.85% saline, Aedes aegypti
ringers (Hayes, 1953) or RPMI tissue culture medium


16
underwent a number of cleavages to increase the surface
area available for merozoite budding. Since they did
not observe the detachment of separate nucleated masses
from the parent schizont, they suggested that the term
" p s eudocy t ome r e11 as described by Garnham (1951) should
be applied to descriptions of co1umbae.
Based on their study of mature merozoites of H.
co1umbae and their comparisons to merozoites of other
haemosporidia, Bradbury and Gallucci (1972) felt that
Ha emoproteu s and P1 a smodium were more closely related
than Leucocytozoon Mature merozoites of both PI asmod i um
and Haemoproteus appear structurally identical, while
those of Leucocytozoon differ in number of limiting
membranes, overall shape, absence of cytostomes and absence
of a mitochondrion associated spherical body (Bradbury
and Gallucci, 1972).
Studies of the sporogonic stages of avian
haemoproteids have been limited to the sporozoites of
H. co1umbae. Klei ( 1972) and Klei and DeGiusti (1973)
failed to find significant differences between the
structure of these sporozoites and those of other
haemosporidians The only u1trastructura 1 study of a
haemoproteid oocyst was conducted by Sterling and DeGiusti
(1974) on metchnikovi a parasite of turtles that is
transmitted by the tabanid fly, Chrysops ca11idus. The
small size of the oocysts they observed as well as the


291
Russell, P.F. 1959. Insects and the epidemiology of
malaria. Ann. Rev. Ent. 4: 415-434.
Russell, P.F., B.N. Mohan and P. Putnam. 1943. Some
obsetvations on spleen volume in domestic fowls in the
course of P1 a smodiurn gal 1 in a c e urn studies. J .
Parasitol. 2T1 203-216.
SAS User's Guide: 3asics. 1982. SAS Institute, Cary, North
Carolina. 921 pp.
SAS User's Guide: Statistics. 1982. SAS Institute, Cary,
North Carolina. 584 pp.
Schorger, A.W. 1966. The wild turkey. University of
Oklahoma Press, Norman. 625 pp.
Schrevel, J., G. Asfaux-Foucher and J.M. Bafort. 1977.
Etude u1trastructura1e des mitoses multiples au Cours
de la sporogonie du Plasmodium b. berghei. J.
Ultrast. Res. 59: 332-35TT ~
Sergent, Ed., and Et. Sergent. 1906. Sur le second hote de
I 'Ha emo proteus ( Halteridium) du pigeon. (Note
prliminaire). C.RT Stances Soc. Biol. Fil. 61:
494-496.
Service, M.W. 1971. Adult flight activities of some
British Cu 1icoi des species. J. Med. Ent. 8: 605-609.
Sibley, L.D., and J.K Werner. 1984. Susceptibility of
pekin and muscovy ducks to Haemoproteus nettionis.
J. Wild!. Dis. 20: 108-113.
Sinden, R.E., and K. Strong. 1973. An u11rastruetura 1 study
of the sporogonie development of P1 a smodiurn falciparum
in Anopheles gambiae. Trans. Roy. Soc. Troo. Med.
Hyg. 72: 477-491.
Smith, H.A. T.C. Jones and R.D. Hunt. 1972. Veterinary
Pathology. Lee and Febiger, Philadelphia. 1521 pp.
Snow, W.E. 1955. Feeding activities of some
b 1 ood-sucking Dptera with reference to vertical
distribution in bottomland forest. Ann. Ent. Soc.
Amer. 48: 512-521.
Sterling, C.R. 1972. U11rastructura 1 study of gametocytes
and gametogen esis of Haemoproteus me t chn i kov i J.
Protozool. 19: 69-76.


Figure 88. Mega 1 oscli i zon t and associated phagocytic cells
(PC) from pectoral muscle of a high dose
bird that died spontaneously at 20 days
post-infection. The wall (\V) of the
megaloschizont is thick and laminated.
Large quantities of amorphous, granular
material are exterior to the wall. A mature
merozoite (Me) is located within the
interior of the cyst. X 26,100.


25
Meteorological data for the study sites at Paynes
Prairie and Fisheating Creek were obtained from the nearest
weather stations, operated on a continuous basis
(Climatological Data: Florida, 1982, 1983, 1984). In
northern Florida, data were obtained from the U.S. weather
station operated at the Gainesville airport, located
approximately 8 km N of the study areas at Paynes Prairie.
In southern Florida, data were obtained from the U.S.
weather station at LaBelle, approximately 24 km SVV of
the study site at Fisheating Creek.
Activity Cycles
Capture times were determined for specimens of C.
edeni, C. hinmani, C. arbor¡cola and C^ knowltoni obtained
in Bennett traps at Paynes Prairie and Fisheating Creek.
Capture times for individual specimens of Cu 1 icoi des were
calculated as the midpoint of the Bennett trap run when
they were collected. The data were plotted by species,
site, year and quarter (January-March, April-June,
Ju1y-September and October-December) as the number of
minutes before or after nautical sunset on the date of
capture. Sunset times were calculated from standard tables
(Nautical Almanac, Nautical Almanac Office, U.S. Naval
Observatory) and adjusted for the proper latitude and
longitude of each study site. Scatter plots for
individuals of each species were very similar when examined


120


163
Table 22. Classification summary of a nearest neighbor
analysis of a set of calibration data composed
of adjusted measurements of host cells
infected with microgametocytes. Discriminant
scores were classified with a discriminant
function derived from the calibration data
set in Tab 1e 21.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
9 (60.0)+
0
(0.0)
3
(20.0)
3 (20.0)
15
(100)
Pheasant
0 (0.0)
7
(46.7)
6
(40.0)
2 (13.3)
15
(100)
Turkey
3 (20.0)
2
(13.3)
9
(60.0)
1 (6.7)
15
(100)
Tota 1
12 (26.7)
9
(20.0)
13
(40.0)
6 (13.3)
4S
(100)
Priors*
0.3333
0
. 3333
0
. 3333
+ Standard Deviation
* Prior probability of being assigned to that class


Figure 71.
Exf1 age 11 ating mi ergametocyte At the same
time the outer layer of the pellicle separates
from the parasite (double arrow), breaks appear
in the inner osmiophilic layer (arrow). X
46,400.
Figure 72. Higher magnification of Figure 71. X 72,000.


116


179
Mesaloschizont s
Mature, 19-day-old mega 1osch¡zonts were extracellular
and surrounded by a thick, laminated cyst wall composed of
electron dense, amorphous material (Figure 88). Host
tissue adjacent to megaloschizonts was necrotic and
infiltrated with phagocytic cells (Figure 88). The region
between healthy muscle tissue and megaloschizonts contained
large quantities of amorphous, granular material with
the same electron density as the cyst wall (Figure 88),
membranous vesicles, free mitochondria with swollen and
ruptured cristae and scattered fragments of myofibrils.
Many mitochondria were opaque and electron dense. The
amo rphous, granular material exterior to the
megaloschizonts appeared to be deposited in layers onto the
outer surface of the cyst wall (Figure 88). Phagocytic
cells adjacent to ruptured megaloschizonts were often
packed with merozoites.
Megaloschizonts contained merozoites that developed as
buds from discrete cytomeres (Figure 89). The
merozoites were bound by a single unit membrane that was
underlaid by an interrupted intra-membranous layer composed
of 2 unit membranes in close apposition to one another
(Figure 89). Merozoites had 3 anterior polar rings, a pair
of electron-dense rhoptries, micronemes, a nucleus bound by
2 unit membranes and a single mitochondrion with tubular
cristae (Figures 89, 90, 91). All merozoites also had a


Figure 14. Transmission vs. abundance of 3 species of Cu I icoi des at
Site B at Paynes Prairie that were able to support
development of Haemoproteus me 1e a g ridis =
Bennett trap catch; OO ~= New Jersey light trap
catch). The kite diagram at the top of the figure
indicates the % of sentinel birds that became
infected with Ha emo proteus me Ie a g rid i s during 4-week
periods between July, 1982 and June, 1984. Blank areas
in the diagram indicate periods where transmission
did not occur. Marks on the x-axis that follow each
month indicate the middle of that month.


Figure 87. Six-day-old oocyst from a specimen of
Cu 1 i coi de s eden i. The oocyst has ruptured,
releasing the mature sporozoites. A
degenerating residual body (R3) containing
lipid-like vesicles (L) and electron dense
masses (arrows) remains. X 26,100.


53
in the biting activity o£ eden i in July and December,
1983, at Site A (Figure 13) and in December, 1982, and May,
1983, at Site B (Figure 14). Transmission did not occur
between May and September, 1982, at either site, in
spite of relatively large, but variable, catches of C.
edeni (Figures 13, 14). Similar anomalies between the
transmission of me 1eagridis to sentinel birds and the
abundance and biting activity of eden i occurred in
October, 1982, and October, 1983, at site B (Figure 14)
when catches of edeni were low, in spite of high rates
of transmission.
Cu 1 icoi de s hinma ni was active only during the
warmer months of the year, between May and October. Both
biting activity and abundance were essentially unimodal in
distribution, with variable peaks between May and
October during periods of active transmission of H.
me 1 eag ridis to sentinel birds. Cu 1 icoi des hinman i wa s
absent during periods of high transmission of H
me 1eagridis in December, 1982, and December, 1983 (Figures
13, 14).
Cu 1 icoi des arbor i col a was absent at both sites in 1983
and 1984 during the cooler winter months of January,
February and March. Peaks in biting activity occurred
in September, 1982, April and September, 1983, and
March, 1984, at each site. Light trap collections during
the same period at Site B had a similar, bi modal


223
subtropical climate of southern Florida to the more
temperate climate of northern Florida.
The high prevalence of transmission of roe 1eagrid i s
to sentinel turkeys at Fisheating Creek was correlated with
the large number of specimens of edeni that were
captured in light traps and 3ennett traps throughout the
year. The repeated isolations of me 1eagridis from pools
of wild-captured, unengorged specimens of O edeni provide
further evidence of the importance of this species as a
major vector of meleagridis. The conspicuous
absence of individuals of hinmani, C. arbor ico1 a and C.
knowlton i at times when transmission was high, their lower
biting
rates and the fa i
lure
to obtai
n wild i s o1 a t
i ons
of the
parasite fr oin i n
d i v j
dual s
o f
these speci
e s ,
indicates that they may
play
only
mi
nor roles in
the
epizootiology of me 1eagrid i s .
The low numbers of specimens of eden i that were
captured during abnormally cool, wet weather in February
and March, 1 983 and November, 1984, and the absence of
detectable levels of transmission of me 1eagridi s may be
related. Greiner (pers. c onm.) also failed to detect
transmission of HL meleagridis to sentinel turkeys in
the same area during abnormally wet weather in March and
April, 1978. It is unclear whether the population of
adult Cu 1 icoi des was inactive because of cool temperatures


Figure 1
Figure 2
Figure 3.
Figure
Figure 5
Figure
Ookinete of Haemoproteus me 1eagr i d i s from the
midgut of a specimen ol Cu 1icoides edeni, 24
hours after the fly engorged on ah infected
turkey. A mass of pigment (arrow) is located
near the posterior end of the organism. Bar = 10
um. Note: Figures 1-6 were taken with Normarski
contrast interference microscopy.
Developing oocysts (arrows) of Haemoproteus
me 1 eagr i d i s on the midgut of a spec ¡men oT
Cu 1 i coi des eden i 4 days after the fly had taken
a blood meal from an infected turkey. Bar = 50
um.
A 6-day-old, degenerating oocyst of Haemoproteus
me 1 e ag ridis from a specimen of Cu 1 icoide¥
edeni. The oocyst contains large retractile
granules (arrow). Bar = 10 um.
A mature, 6-day-old oocyst of Haemoproteus
me 1eag rid i s from a specimen of Cu 1 i coi des
e d e n i The oocyst is packed with slender
sporozoites that are parallel to one another. A
small, retractile residual body (arrow) is
present. Bar = 10 um.
One of the 2 salivary glands from a specimen of
Cu 1 i coi de s e d e ni that had engorged on an
infected turkey 7 days earlier. The gland
consists of an elongate primary lobe and several
smaller, apical lobes. The shadows of
sporozoites are barely visible in secretory
cells composing the primary lobe (arrows). Bar
= 10 um.
A crushed salivary gland from a specimen of
Cu 1 icoi de s ed e ni that had engorged on an
infected turkey 7 days earlier. Numerous
elongate sporozoites (arrows) are located within
the secretory cells. Bar = 10 um.


Figure 22. An intact mega 1osch i zont from the pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infection .
Node-like constrictions (arrows) occur along
its length. Hematoxylin and eosin. Bar =
1 OOum.
Figure 23. An intact mega 1 oschizont from pectoral
muscle of a high dose bird that died
spontaneously at 19 days post-infect i on.
The mega 1 oschizont is surrounded by a thick,
hyaline wall (arrows). The interior is packed
with minute merozoites. Cytomeres are not
evident. The megaloschizont is surrounded by a
mixed inf1anrnatory infiltrate composed of giant
cells (double arrow), macrophages, heterophils
and red blood cells. Hematoxylin and eosin.
Bar = 50 urn.


Figure Page
54 Swollen, hyaline and disrupted pectoral muscle
fibers from a turkey with a 5-day-old
experimental infection of Haemoproteus
me 1 eagr i d i s 148
55 Regenerating muscle fibers from pectoral
muscle of a turkey with an 8-day-old
experimental infection of Haemoproteus
me 1 eagr i d i s 148
56 Deteriorating 17-day-old mega 1oschizont 148
57 Megaloschizonts from pectoral muscle of a
naturally infected Wild Turkey 150
58 Mega 1oschizont from pectoral muscle of a
naturally infected Wild Turkey 150
59 Parasitemias per 10,000 red blood cells for
turkeys, the Chuckar and the Ring-necked
Pheasant with experimental infections of
Haemoproteus me 1eagridi s 171
60 Macrogametocyte of Haemoproteus me 1eagridis
from an experiment ally infected turkey 777.... 173
61 Microgametocyte of Haemoproteus me 1eagridis
from an experimentally infected turkey 173
62 Macrogametocyte of Haemoproteus me 1eagridis
from an exper¡menta 11 y infected Chuckar TT.... 173
63 Microgametocyte of Haemoproteus me 1eagridis
from an experimentally infected Chuckar TT.... 173
64 Macrogametocyte of Haemoproteus me 1eagrid i s
from an experimentally infected Ring-necked
Pheasant 173
65 Microgametocyte of Haemoproteus me 1eagridis
from an exper¡mental 1y infected Ring-necked
Pheasant 173
66 Circulating microgametocyte 182
67 Circulating macrogametocyte 184
68 Higher magnification of Figure 67 184
x i v


61
Table 6. Susceptibility of wild-caught specimens of
Cu 1icoi des to Haemoproteus me 1eagridis Fractions
represent numbeT pos itive/ number examined and are
followed by percent of total.
Development of Parasite
Spec ies
None
Part
i a 1 *
Complete**
c
eden i
32/52
(61.5%)
c
arboricol a
6/28
(21.4%)
o
haematopotus
1/6
(16.7%)
c
hinma ni
8/72
(11.1%)
c
knowltoni
1/14
(7.1%)
c
paraensis
1/2
(50.0%)
o
nanus
8/27
(29.6%)
c
scan 1 oni
3/25
(12.0%)
c
bauer i
2/24
(8.3%)
c
crepuscularis
3/3 (100%)
**
Degenerating oocysts present
Invasion of salivary glands by sporozoites


LOG|q (Parasitemia +1)
O rv> w ^
Oil
Crisis


174
macrogainetocytes, the endoplasmic reticulum contained a
moderately dense, amorphous material (Figure 68). The
endoplasmic reticulum was occasionally continuous with the
innermost, osmiophilic layer of the pellicle (Figure
68). Ma c r ogame t oc y t e s also contained membrane-bound,
osmiophilic bodies (Figure 67).
Gametocytes contained a single cytostome,
surrounded by 2 electron dense rings (Figure 69). Food
vacuoles containing host cell cytoplasm and limited by the
2 outer membranes of the pellicle formed at the inner
surface of the cytostome, between the 2 electron dense
rings. Other food vacuoles in the macrogametocyte were
bound by a single unit membrane and contained
osmiophilic masses of pigment (Figure 69).
Gametogenes i s
In the earliest stages of gametogenes i s, gametocytes
rounded-up within their host cells (Figures 70, 71,
74). The outer unit membrane of the 3-Iayered pellicle
separated from the parasite and formed membranous vesicles,
coils and whorls or remained free as discrete fragments in
the red blood cell cytoplasm (Figures 70, 72, 73).
Dense granules, approximately the same size and electron
density as free ribosomes, occasionally lined the inner
side of the membrane in a single layer (Figure 73). The
thickened, osmiophilic inner layer of the pellicle of


230
Pathogen icity
Exoerythrocytic Development
The description of the schizonts of columbae by
A r a g a o (1908) was the first detailed account of the
exoerythrocytic development of any haemosporidian parasite
(Garnham, 1966). In the 77 years since then, little
more has been learned about the exoerythrocytic development
of avian haemop r o teids Because experimental infections
have been difficult to perform for most species of
Haemoproteus, virtually nothing is known about their early
stages of exoerythrocytic development.
The results of this study demonstrate that H.
me 1eagridis undergoes at least 2 generations of schizogony
in skeletal and cardiac muscle of experimentally
infected turkeys. Following i ntraperitonea 1 inoculation,
sporozoites probably gained access to the circulation
via the lymphatic system and became localized in the
rich capillary beds of the skeletal muscle. It is unclear
whether they began their development within capillary
endothelial cells, satellite cells or myofibroblasts. The
early schizogonic stages of Sarcocy s t i s may occur in all 3e
cell types, as well as macrophages, but are most corrmon in
endothelial cells (Entzeroth, 1983; Dubey et al., 1983;
Cawthorn et al., 1983; Leek et al., 1977). The early
schizogonic stages of other cyst-forming sporozoans,


232
8 i ossum-Headed Parakeets, naturally infected with H.
desser i. They noted that development appeared to be within
muscle fibers, rather than cells of the reticuloendothelial
system. In their study, as well as the one by Fanner
( 1965), the mega 1 os ch i zon t s were too large to precisely
determine the type of host cell.
Mature, 5-day-old, first-generation schizonts of H.
me 1eagridis contained elongate zoites that differed in size
and morphology from the pre-erythrocytic merozoites of
mature, 17-day-old, second-generation mega 1 osch i z on t s
(Figures 44, 50). They were approximately the same size as
first-generation hepatic schizonts of simo n di but
lacked the hypertrophied host cell nucleus that is
characteristic of Leucoc y t ozoon infections (Huff,
1942). In addition, first-generation merozoites of
Leucocytozoon are oval to round in shape, rather than
elongate (Desser, 1967, 1974; Akiba et al., 1971; Huff,
1942). The first-generation schizonts of H^ me 1eagridis
were more similar in morphology to the first-generation
schizonts of Sarcocystis (Leek et al., 1977; Dubey et al.,
1983). The small size of schizonts found in the 8-day-old
infection suggests that the first-generation schizonts
ruptured between 5 and 8 days post-infection and
released zoites that initiated a second-generation of
schizogony.


54
distribution. A bimodal distribution was less evident
in light trap collections from Site A (Figures 13, 14).
Biting activity of a r boricol a was very low or absent
during periods of high transmission of me 1eagridis
between November and December, 1982, and between October
and December, 1983.
Fisheating Creek. Bennett trap and light trap
collections of edeni, C. h i nmani and CL arbor ico1 a from
Fisheating Creek had seasonal patterns that were similar to
catches from Paynes Prairie. Cu 1 icoides edeni remained
active throughout the year, with fewer numbers and lower
biting activity from December, 1982, through March, 1983,
and January and November, 1984, during cooler winter
weather (Figure 18). Ha emopr o t e u s me 1eag ridis was
transmitted at high levels to exposed sentinel birds
between May, 1983, and October, 1984.
Cu 1 icoi des h i nmani and (L arboricola were absent or
present in low numbers from December, 1982 March, 1983,
and from December, 1983, and March, 1984, when average
monthly temperatures were lowest (Figures 17, 18).
Light trap and Bennett trap collections of hinmani were
bimodal in distribution in 1983 with peaks in May and
September. The distribution was more unimodal in 1984.
Light trap collections of arborico1 a were variable
throughout 1983 and 1984, but were essentially unimodal


1 55
cells from the turkey underwent the greatest enlargement.
Cells from all 3 hosts also had atrophied nuclei with
smaller lengths, widths and areas than nuclei from
unparasitized red blood cells (Table 18). Infected host
cells from the Ring-necked Pheasant underwent the greatest
nuclear atrophy.
A discriminant analysis was performed on adjusted
variables from a calibration data set composed of
measurements of 15 Chuckar cells infected with
ma crogametocytes 15 infected turkey cells and 7
infected pheasant cells. It correctly identified 73.3% of
the Chuckar scores, 66.7% of the turkey scores and none of
the pheasant scores. Three (42.9%) of the pheasant scores
were identified as Chuckar and 3 (42.9%) were identified as
turkey percentages approximately equal to what would
be expected by chance alone (Table 19). One pheasant score
failed to meet the criteria for inclusion in any of the
3 categories.
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 host cells from
the Chuckar and turkey and 3 host cells from the pheasant.
Three (75%) of the Chuckar scores were classified as turkey
and 2 (66.7%) of the pheasant scores were classified as
Chuckar. Two (50%) turkey scores were correctly
identified, but the classification was only slightly
less than would be expected by chance alone (Table 20).


Table 11. Average organ* weights at necropsy
Group
N
Heart
(p = 0.112)
Liver
(p = 0.0641)
Spleen
(p=0.0002)
Control
1 1
0.443
1.70
0.093
Low
12
0.457
1.74
0.130
High
8
0.489
1.93
0. 163
Weights expressed as % of total body weight at necropsy


Figure 42. Three-day-old schizont (arrow) from pectoral
muscle of an experimentally infected
turkey. The schizont is within a
capillary. Hematoxylin and eosin. Bar = 10
um.
Figure 43. Five-day-old schizont from pectoral muscle
of an experimentally infected turkey. The
schizont is located between muscle fibers
and is packed with dark granules. Hematoxylin
and eosin. Bar = 10 um.
Figure 44. Five-day-old schizont from pectoral muscle
of an experimentally infected turkey. The
schizont is located within a muscle bundle and
is packed with elongate zoites. Hematoxylin
and eosin. Bar = 10 um.
Figure 45. Eight-day-o 1d mega 1 oschizont (arrow) from
pectoral muscle of an experimentally
infected turkey. The megaloschizont is located
within a capillary and contains several
dark-sta in i ng nuclei. Hematoxylin and eosin.
Bar = 10 um.
Figure 46. Eight-day-o1d mega 1 oschizont from pectoral
muscle of an experimentally infected turkey. A
giant cell (double arrow) is adjacent to the
mega 1 oscli i zon t (arrow). Hematoxylin and eosin.
Bar = 10 um.
Figure 47. Fourteen-da y-o 1d me ga 1 oschizont from
pectoral muscle of an experimentally
infected turkey. The mega 1oschizont is located
within a muscle fiber and is surrounded by a
thick, hyaline wall (arrow). Hematoxylin
and eosin. Bar = 10 um.


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS i i i
LIST OF TABLES ix
LIST OF FIGURES xi
ABSTRACT xvii
INTRODUCTION I
Epizooti o Iogy 6
Pathogenicity 9
Hos t Spec ificity II
Fine Structure 13
Objectives 17
MATERIALS AND METHODS 20
Ep izootiology 20
Sentinel Study 20
Vectors 21
Activity Cycles 25
Experimental Infections 27
Estimation of Prevalence 29
Pathogen icity 30
Experiment 1 Pathology 30
Experiment 2 Exoerythrocytic Stages 34
Host Spec ificity 36
Infection of Hosts 36
Morphometric Analysis 37
Fine Structure 39


15
a dense ring anterior to the ribs rather than from a polar
ring, as is true of most sporozoans. Additional studies
of the ookinetes of ve 1ans or other haemoprote i ds
transmitted by Cu 1 icoi des are needed to determine whether
these differences are important.
Few studies of the schizonts and sporogonic stages
of avian haemoproteids have been conducted. Bradbury
and Gallucci (1971, 1972) examined the fine structure
of schizonts and differentiating merozoites of H.
co1umbae. They clarified a number of inconsistencies
and errors made by Aragao (1908 ) in his description of
schizonts from lung tissue in a Rock Dove. Bradbury and
Gallucci (1972) failed to find evidence of sexual
dimorphism in the schizonts they examined and suggested
that the differential staining noted by Aragao (1908)
was probably due to differences in stage of development.
Desser (1970a) described a conspicuous sexual dimorphism
in merozoites of Leucocy tozoon s imond i but Bradbury and
Gallucci (1971) failed to find similar differences among
mature merozoites of columbae
Bradbury and Gallucci (1971) also clarified the use
of the term "cytomere". Aragao (1908) and later Bray
(1960) used the term to refer to development of separate,
uninucleate masses that eventually underwent extensive
nuclear division to produce merozoites. Bradbury and
Gallucci (1971) found that the schizonts of H. coIumbae


at sotting Cu 1icoi des. John Bogue deserves recognition for
his help with holding and bleeding birds and for his unique
sense of humor.
Finally, 1 would like to reserve a special
acknowledgement for my fellow graduate student, Lora G.
Rickardfor help with catching Cu 1 icoi des, holding turkeys
and swatting mosquitoes at Fisheating Creek and Paynes
Prairie, for all the stimulating discussions about worms
and protozoans and, most of all, for her friendship.
This research was funded in part by the National Wild
Turkey Federation and grant number 1270-G from the Florida
Game and Freshwater Fish Conmission.


8S
MONTHS


265
polarization of mi c r ogame t ocy t e s of me 1eagridis into
2 halves was not observed. Atypical centrioles were not
detected, but they may have been missed since serial
sections were not cut.
Microgametes of me 1easridis were similar to
microgametes of other species of Haemoproteus .
Microgametes of most other species contain a single axoneme
and a small, membrane-bound nucleus. Mitochondria have not
been observed (Aikawa and Sterling, 1974a; Sterling, 1972;
Desser, 1972a). Bradbury and Trager (1968a) observed 2
axonemes in mi c r ogame t e s of co 1 umbae but these
observations have not been confirmed (Aikawa and Sterling,
1974a). The periodic striations observed in the central
microtubules of axonemes of me 1eagridis have been
reported in axonemes of me t c h nik o v i and si mo n di
(Sterling, 1972; Aikawa et al., 1970).
Oocyst s
Most ultrast ructuraI studies of the oocysts of
haemosporidian parasites have been limited to avian and
mammalian species of PI a smodium (Mehlhorn et al., 1980;
Sinden and Strong, 1978; Canning and Sinden, 1973; Howells
and Davies, 1971; Terzakis, 1971; Terzakis et al., 1966;
Duncan et al., 1960) and to 3 species of Leucocy tozoon
Oes ser and Allison, 1 979; Wong and Desser, 1 976 ;
Desser, 1972c). In spite of the importance of comparative


48
for specimens of e d e n i C a r bo cicola and C.
know 1 toni were less clear and varied from quarter to
quarter. At Fisheating Creek, mean capture times for
specimens of arborico1 a and knowltoni were not
significantly different during any quarter. However,
the peak biting activity for specimens of edeni was
significantly earlier than specimens of arboricola
during quarters 2 and 4, but not during quarters 1 and 3.
At Paynes Prairie, specimens of edeni had a peak capture
time that was significantly earlier than specimens of C.
arboricola during quarters 3 and 4, but not during quarter
1. During quarter 2, specimens of eden i had a peak
in activity that was significantly later than specimens of
C. arbor ico1 a (Table 7).
In spite of these differences, the same trend was
evident at each site. Specimens of hinmani reached a
peak in biting activity before sunset. They were followed
from 0 to 19 minutes after sunset by specimens of C.
edeni. Specimens of arboricola were most active from 8
to 55 minutes after sunset. Specimens of knowltoni were
the last to become active between 26 and 51 minutes
after sunset. When mean capture times for individuals
of each species are compared between Paynes Prairie and
Fisheating Creek, differences are usually within 1 standard
deviation of each other.


100
Microscopic observations spontaneous deaths. The
most significant lesions found in tissues from the 4
spontaneous deaths were in the skeletal muscles. All 4
birds had numerous intact and degenerating, fusiform
me ga 1 oschizonts between the muscle fibers. The
schizonts ranged from 43 to 155 urn in diameter with a mean
diameter of 95.0 urn (n = 35, SD = 34.67). Depending on the
plane of section, they ranged up to 600 urn in length. Most
mega 1 oschizonts were oriented parallel to the muscle
fibers, although a few were perpendicular. Many were
outlined by a thick, hyaline wall which occasionally
indented slightly to form node-like constrictions (Figures
22, 23).
Mature mega 1 oschizonts were packed with small,
spherical merozoites less than 1 urn in diameter. Each
contained a small mass of chromatin. Some immature
mega 1 oschizonts contained irregularly shaped cytomeres
of different sizes that were located next to the cyst
wall. Chromatin masses were arranged around the periphery
of each cytomere. Other mega 1oschizonts were packed
with round, granular, dark blue cytomeres. As
mega 1oschizonts matured, the number of cytomeres increased
in inverse proportion to their diameter. Merozoites
developed as buds from the outer surface of the cytomeres.
A severe hemorrhagic myositis was associated with the
mega 1 osch i zonts Ruptured schizonts were surrounded by


124


159
Table 13. Classification summary of a nearest neighbor
analysis of a set of calibration data
composed of adjusted measurements of
macrogametocytes Discriminant scores were
classified with a discriminant function derived
from the calibration data set summarized in
Table 12.
Classified Into Species
Spec ies
Chuckar
Pheasant
Turkey
Other
Total
Chuckar
12
(80.0)
+ 0
(0.0)
2
(13.3)
1 (6.7)
15
(100)
Pheasant
0
(0.0)
1
(14.3)
5
(71.4)
1 (14.3)
7
(100)
Turkey
2
(13.3)
0
(0.0)
12
(80.0)
1 (6.7)
15
(100)
Total
14
(37.8)
1
(2.7)
19
(51.4)
3 (8.1)
37
(100)
Priors*
0.
,4054
0.
,1892
0
.4054
+ Percent of total
* Prior probability of being assigned to that class


160
Table 14. Classification summary of a nearest neighbor
analysis of a set of test data composed of
adjusted measurements of macrogametocytes. The
discriminant function derived from data in Table
12 was used to classify the discriminant scores.
Classified into Species
Spec i es
Chuckar
Pheasant
Turkey
Other
Tota 1
Chuckar
1 (25.0)
+ 0 (0.0)
3 (75.0)
0
(0.0)
4
(100)
Pheasant
1 (33.3)
I (33.3)
1 (33.3)
0
(0.0)
3
(100)
Turkey
0 (0.0)
0 (0.0)
4 (100)
0
(0.0)
4
(100)
Total
2 (13.2)
1 (9.1)
8 (72.7)
0
(0.0)
11
(100)
Priors*
0.4054
0.1892
0.4054
+ Percent of total
* Prior probability of being assigned to that class


243
zoites and small, spherical pre-erythrocytic merozoites
were found in the spleen of I high dose bird that died
spontaneously. Other studies have reported the development
of exoerythrocytic schizonts of PI a s m o d i jum and
Leucocytozoon in reticular cells of the spleen (Huff, 1969;
Akiba et al., 1971). Host cell nuclei of the reticular
schizonts were not hypertrophied as is characteristic of
Leucocytozoon infections.
The reductions in growth and weight gain in
experimentally infected birds were dose dependent and most
pronounced between 1 and 3 weeks post-infection,
during development of second-generation
mega 1 oschizonts (Figures 37, 38). The onset of lameness
and anorexia in the high dose birds was approximately 1
week later than in the earlier series of experimental
infections and was probably associated with the
infla mm atory response to the second-generation
mega 1oschizonts. Because high dose birds received fewer
than 1/3 as many sporozoites as birds that were infected to
study exoerythrocytic development, pathological changes
associated with development of the f i rst-gene r a t i on
schizonts may not have been as severe.
Following the crisis, all infected birds improved.
This was most evident among turkeys in the high dose group
that exhibited little significant weight gain between weeks
0 and 2, between weeks 1 and 3 and between weeks 3 and 4


16
15
14
13
12
II
10
9
8
7
6
5
4
3
2
I
0
Culicoides knowltoni
00

X = 28.3
0
#
0 0
0 0
00 00 m 004
LJL
j i i i
140 -120 -100 -80 -60 -40 20 0 20+ 40 60 80 100 120 140
CAPTURE TIME


Figure 12. Modified New Jersey suction trap catches of
specimens of C u1 ic o i d e s hinma ni a t
Fisheating Creek dur ing the Ma r cE~¡ April and
May, 1 983 collecting trips. (9 = canopy
trap; OO = ground trap). Rising and setting
suns indicate the 2-hour sampling period
that included dawn or dusk.


235
Garnharn (1951) to describe them. Bray (1960) attempted to
resolve differences between the terms "cytomere" and
"pseudo-cytornere" and simplify terminology by redefining
cytomeres as "separate growing nucleated masses,
produced within a process of schizogony by the division of
a schizont.". He stated that the process "may produce
further cytomeres but produces uninucleated organisms as an
end point." Unfortunately, the redefinition has not
been widely accepted. Most authors have applied the
term "pseudo-cytomere" to descriptions of schizonts that
contain mu 11 i nuc1eated masses of protoplasm that remain
attached to the parent schizont (Bradbury and Gallucci,
1972; Sterling and DeGiusti, 1972). Since knowledge of the
entire developmental sequence is required to determine
whether a mu 1 t i nuc 1 ea t ed mass developed from a separate
uninucleate body or from a multinucleated mass that
detached late in development from the parent body, use
of either term may be misleading when the complete
morphogenesis is unknown.
During the development of second-generation
mega 1 oschizonts of meleagridis, cytomere formation
occurred between 8 and 14 days pos t-i n f ect i on At the
present time it is not possible to determine whether the
cytomeres developed by segmentation of the mu I tinuc1eate
8-day-old schizonts into uninucleate masses or by
detachment of multinucleated masses from the parent body.


17
formation of sporozoite buds from a single residual body
closely resembled oocyst development described by light
microscopy for avian haemoproteids transmitted by
Cu 1icoi de s (Falls and Bennett, 1960; Kahn and Falls,
1971).
Object i ves
Ha emop r oteu s me 1eag rid i s was first reported in a
domestic turkey, Me 1eagris ga11opavo, in Texas by Morehouse
( 1945). Since then, this parasite has been found in wild
and domestic turkeys throughout the Nearctic range of
the host and in Venezuela (Greiner and Forrester, 1980).
Eve et al. (1972) and Eve et al. (1972) speculated that
H, me 1eag ridis might be a potential pathogen of wild and
domestic birds. Much of the evidence for this was circum
stantial, at best, and based on observations that high
parasitemias seemed to coincide with periods of high
mortality in wild birds.
The existence of a domesticated host that is
inexpensive and available throughout the year has made
H. me 1eagridis an attractive laboratory model for the
study of avian haemoproteids. Until recently this model
has not been feasible because the vectors of hU me 1eagridis
have been unknown. In a survey of ectoparasites from


WEEKS


246
Leek et al. ( 1977) observed a significant drop in total
serum protein in lambs infected with Sa rcocys tis The
decrease occurred during development of first-generation
schizonts in endothelial cells. Leek et al. suggested that
the drop in serum protein levels resulted from
glomerulonephritis associated with development of schizonts
in endothelial cells of the kidneys. Similar lesions were
not detected in infected birds examined in this experiment.
The increase in plasma protein concentrations in
the high dose group at 2 weeks pos t-i nf ect ion may have been
related to dehydration observed among these turkeys (Figure
40). Augustine (1982) noted similar increases in plasma
protein concentrations when turkeys were deprived of water
for periods of up to 72 hours. Wien birds were deprived of
both food and water, significant changes in plasma protein
concentrations did not occur. Since feed consumption
was not measured in the pathogenicity experiment, it is
unclear whether the high dose birds ceased eating and/or
drinking completely. The diarrhea associated with the
concurrent S a 1 inon ella infection may have acted in
conjunction with decreases in food and water consumption to
dehydrate the birds and concentrate total plasma proteins.
The increase in average plasma protein
concentration at 5 and 6 weeks post-infection may
reflect the synthesis of immunoglobulins (Figure 40).
Other studies of avian species of P1 a smodium and


47
descriptions of neotypes of me Ieagridis (Greiner and
Forrester, 1980).
Act ivity Cycles
Bennett trap catches. Of the 5 species of Cu 1 icoi des
capable of supporting development of me 1 eagridis,
individuals of only 4 species, O edeni, C. hinma ni C.
a r borico1 a and knowltoni were captured in sufficient
numbers to permit analysis of their times of capture. A
scatter plot of capture times for specimens of C.
hinmani from 2.5 hours before sunset to 2.5 hours after
sunset was unimodal with a peak at 41.6 minutes before
sundown (Figure 7). Scatter plots of capture times for
specimens of edeni, C. arboricola and O knowltoni were
also unimodal, but average peaks were 10.3, 19.2 and
28.3 minutes after sundown, respectively (Figures 8, 9,
10).
Comparisons of mean capture times among individuals of
the 4 species revealed the same trends at each site (Paynes
Prairie and Fisheating Creek) and during each quarter
(January March, April June, July September, October -
December) (Table 7). Specimens of C^ hinmani had a peak in
biting activity from 24 to 59 minutes before sunset that
was significantly earlier than individuals of the other
species. Significant differences among mean capture times


253
chickens reacted as well to their protein counterparts
from turkeys as they did to their protein counterparts from
Ring-Necked Pheasants. They briefly reviewed other
evidence from hybridization experiments, chromosome
studies, electrophoretic and inmunolog i ca I experiments and
anatomical studies that supported the similarities between
turkeys and other Phasianids. Since cell penetration
by some sporozoan parasites may be mediated by highly
specific cell surface receptors (Miller et al., 1978), the
parasites can, in a sense, be considered as highly specific
probes. The successful experimental transmission of H.
meleagridis from a turkey to a Ring-necked Pheasant and
a Chuckar suggests that all 3 species may have similar cell
surface receptors. Whi 1e these similarities may be of
minor taxonomic significance, they support the data showing
close biochemical similarities among these species.
The family Phasianidae presently contains members of 4
former avian families. Prior to the revision, Bennett
et al. (1982) reported 8 valid species of Haemoproteus
from the family Phasianidae c h u c ka r i rL chap ini,
H. ammo pe r d i s f-K santosdiasi bamb s i co 1 ae HL
1ophor t yx, H. or a ta s i and rK ri 1e yi 2 valid species of
Haemoproteus from the family Numididae pratas i and I-L
si 1 va i 2 valid species from the family Tetraonidae H.
mansoni (= rL canachites) and rK s t ab 1 e r i and 1 valid
species from the family Me 1eagr i didae me Ie ag ridis.


156
Infected host cells mi crogametocytes Host cells
infected with mi ergame toeytes underwent several common
changes in each host species. All had an increase in
average cell length, width and area and a corresponding
decrease in nucleus length, width and area (Figures 61, 63,
65). A lateral displacement of the host cell nucleus
occurred in parasitized red cells from each host. The
d i s pI ac erne n t was greatest in host cells fr om the
Ring-necked Pheasant and least in host cells from the
Chuckar (Table 21). Differences in infected host cell
morphology were most evident in cells from the
pheasant. Infected red blood cells were rounder than those
from the other host species and the red blood cell nucleus
was often displaced to the outer margin of the host cell.
A discriminant analysis was performed on a calibration
data set composed of adjusted measurements of 15 host
cells, infected with mi crogametocytes, from each host
species. It correctly classified 60% of the Chuckar
scores, 46.7% of the pheasant scores and 60% of the turkey
scores (Table 22) .
The derived function was tested with a smaller data
set composed of adjusted measurements of 4 host cells from
each of the 3 host species. Two (50%) of the Chuckar
scores were classified as Chuckar, 1 (25%) of the turkey
scores was classified as turkey and 4 (100%) of the


50
Sentinel Study
Paynes Prairie. Between May, 1982, and July, 1984, 30
of 327 (9.2%) sentinel poults at Site A and 32 of 140
(22.9%) sentinel poults at Site B became infected with
H, me 1 eagr i d i s At Site A, 6 of 166 (3.6%) exposed on the
ground vs. 24 of 161 (14.9%) exposed in the canopy
developed patent infections. A 2 by 2, Chi Square test of
the independence of exposure height and transmission was
highly significant (p<0.01).
During 1982, transmission of me 1eagridis began
in mid-August at Site B and m i d-S ep t erab e r at Site A
(Figures 13, 14), peaked from mid-October to
mid-December during periods of above average
temperatures for northern Florida (Figure 15) and
tapered off at the end of December at Site A and in
mid-January at Site B, with the onset of cooler winter
weather in January, 1983 (Figures 15,17). As average
monthly temperatures reached and exceeded 60o F (Figure
17), transmission began again in mid-April, 1983, at Site B
and early May, 1983, at Site A and continued at both
locations throughout the summer and fall until the onset of
cooler, winter weather in mid-December, 1983 (Figures
IS, 17). Deviations from the average monthly rainfall at
Paynes Prairie were minor throughout most of the study.
Rainfall was above average during March, April, June and


134


Haemoproteus me 1e a g r i d i s was transmitted
experimentally to a Chuckar and a Ring-necked Pheasant, but
not to chickens, Guineafowl or Northern Bobwhites.
The fine structure of circulating and
ex f 1 age 11 a t i ng gametocytes was similar to that of other
avian haemoproteids. The fine structure and development of
oocysts was similar to oocysts of species of
Leucocytozoon Mature megaloschizonts differed
u 11rastructura 11y from similar forms reported from species
of Leucocytozoon
x i x


4
Table 1: Proven and presumed vectors of avian haemoproteids
Spec ies
Vector
Author
H. columbae
H. sacharovi
HT macea 1 1 umi
HT lophor tyx
H. pa 1umbis
HT desseri~
H. ne11ionis
H. velans
H. canachi tes
H. f ringi 1 1ae
H. dan i 1ewskyi
H. me 1eagridis
Pseudolynchi a caar iens i s
P. brunnea
FT capensis
Mi crolynchi a pus i 1 1 a
P. caar iens i s
FT caar iens i s
ST i 1 borne topa ~Tmpres sa
Lynch i a hirsuta
Orn i thorny i a av i cubr i a
Cu 1 icoi des nubeculosus
o
c.
downe si
s ti 1obezziodes
c.
sphagnumensis
C. sphagnumensis
C. crepuscularis*
C.
s ti 1obezziodes*
C.
crepuscular i s*
C.
s ti 1obezziodes*
c.
sphagnumensis*
c
eden i
hinma ni
C.
arboricol a
Sergent and
Sergent, 1906
Aragao, 1908
Gonder, 1915
Aragao, 1916
Huff, 1932
Huff, 1932
O'Roke, 1930
Tar shis 1955
Baker, 1966b
Mi 1tgen et al ,
1981
Fa 1 Iis and Wood,
1957
Kahn and Fa 11is,
1971
Kahn and Fa 11is,
1971
Falls and
Bennett, 1960
Fa 1 1 i s and
Bennett, 1961
Fa 1 1 i s and
Bennett, 1961
Bennett and
Fall is, 1960
Bennett and
Falls, 1960
Falls and
Bennett, 1961
Atkinson et
al., 1983
Atkinson et
al., 1983
Atkinson et
al., 1983
Transmission not carried out, but able to support
development of oocysts and sporozoites.


173
1
*
^60

A. A
# 1
'fW
61
,1*.
62
J
1 *
63J
' *
* 0 \
64 1
11
R ^65


150


263
surrounded 2 central microtubules during the earliest
stages of exf1 age I 1 at i on These were located free in
the cytoplasm in various stages of assembly. One end of
developing axonemes was usually associated with a dense
plaque in the nuclear membrane that usually had associated
intranuclear tubules. Sterling (1972) reported an atypical
centriole embedded in electron dense material adjacent
to the plaque. He observed the attachment of axonemes
to basal bodies that v/ere associated with the atypical
centriole. Aikawa and Sterling (1974a) found that the
intranuclear microtubules extended across the nucleus of
ex flagellating gametocytes of co1umbae to another
electron dense plaque on the opposite side of the
organelle. They observed the condensation of electron
dense material around the base of the plaques and suggested
that this material was eventually incorporated into the
microgamete nucleus. As axonemes began to bud from the
exterior of microgametocytes Aikawa and Sterling
(1974a) described the protrusion of a portion of the
mi c r ogame t ocy t e nucleus to the base of the bud. They
presented micrographs indicating that a portion of the
nucleus, still bound by a nuclear membrane, detached
from the mi ergametocyte nucleus and became incorporated
into the exf1 age 11 ating microgamete as a spiral around the
axonerne. Sterling (1972) described a similar process
during the ex f 1 age 1 1 a t i on of me t chn i Itov i .


280
Bray, R.S. 1960. Observations on the cytology and
morphology of the mammalian malaria parasites. 1. A
process of apparent plasmotomy in the preerythrocytic
phase of Laverania falciparum. Riv. Parassit. 21:
267-276.
Bray, R.S., and P.C.C. Garnham. 1962. The
Gi emsa-co1ophonium method for staining protozoa in
tissue sections. Ind. J. Malar. 16: 153-155.
Canning, E.U., and M. Anwar. 1968. Studies on meiotic
division in coccidial and malarial parasites. J.
Protozool. 15: 290-298.
Canning, E.U., and R.E. Sinden. 1973. The organization
of the ookinete and observations on nuclear
division in oocysts of PI a smodium be rghei .
Parasitology. 67: 29-40.
Cawthorn, R.J., A.A. Gajadhar and R.J. Brooks. 1 984.
Description of Sa rcocys tis rauschorum sp.n. (Protozoa:
Sarcocystid a e ) with experimental cyclic
transmission between varying lemmings (Pi cros tonyx
richardsoni) and snowy owls (Nyctea scandiaca).
Can.J. Zoo 1. 62: 217-225.
Cawthorn, R.J., G.A. Wobeser and A.A. Gajadhar. 1983.
Description of Sarcocystis campes t ris sp.n. (Protozoa:
Sarcocystidae) : a parasite ol the badjer Tax i dea taxus
with experimental transmission to the Richardson's
ground squirrel, Spermophilus ri chardson i i Can.
J. Zool. 61: 370-377.
Chernin, E. 1952 The epi zooti o 1ogy of Leucocy tozoon
si mo nd i infections in domestic ducks Tn northern
Michigan. Am. J. Hyg. 56: 39-57.
Climatological Data: Florida. 1982. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.
Climatological Data: Florida. 1983. National Oceanic and
Atmospheric Administration, Environmental Data and
Information Service, National Climatic Center,
Asheville, North Carolina.


Figure 8. Scatter plot of capture times for specimens of
Cu 1 icoi des edeni that were capture in Bennett traps
at Paynes Prairie and Fisheating Creek. Capture time is
plotted as minutes before or after nautical sunset
(reference line). Arrow indicates mean capture time.


Page
Morphometric Analysis 152
Macrogametocytes 152
Mi ergametocytes 153
Host cells macrogametocytes 154
Host cells microgametocytes 156
Fine Structure 157
Mature Gametocytes 157
Gametogenesis 174
Oocysts 176
Three-day-old oocysts 176
Six-day-old oocysts 177
Megaloschizonts 179
DISCUSSION 213
Epizooti o 1ogy 213
Vectors 213
Sporogonic Stages and Transmission 215
Activity Cycles 217
Transmission and Vector Abundance 220
Pathogenicity 230
Exoerythrocyt ic Development 230
Pathology 242
Host Specificity 250
Fine Structure 257
Mature Gametocytes 257
Gametogenesis 259
Macrogametogenesis 261
Mi crogametogenesis 262
Oocysts 265
Differentiation of the oocyst 266
Nuclear divisions 269
Crystalloid 272
Mega 1oschizonts 273
LITERATURE CITED 276
BIOGRAPHICAL SKETCH 294
v i i