rfiT—WtéTS

LIBKAR?

’ Michigan State-
University

w.—

   
 

This is to certify that the

dissertation entitled

PHYLOGENY AND HISTORICAL BIOGEOGRAPHY
OF THE SCHISTOSOMATIDAE

presented by

Allan Christopher Carmichael

has been accepted towards fulfillment
of the requirements for

fill—degree in M

 

Date 8/9/84

MSUI': an Affirmative Action/Equal Opportunity Inxlimlion

0—12771

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

MSU

LIBRARIES
.4“.

  
   
   

  

A DISSERTATION

Department of Zoology

1984

 

 

 

 

@ Copyright by

ALLAN CHRISTOPHER CARMICHAEL

1984

ii

 

 

ABSTRACT

PHYLOGENY AND HISTORICAL BIOGEOGRAPHY
OF THE SCHISTOSOMATIDAE

by

Allan Christopher Carmichael

Phylogenetic relationships of blood flukes of the
trematode family Schistosomatidae were assessed and compared
with host phylogenies in order to examine alternative
hypotheses concerning coevolution and host shifting during
phylogenetic history. Coevolution is suggested by con-
gruence between host and parasite cladograms while pat-
terns other than strict co-occurrence may indicate that
host shifting occurred at some point in an organism's
lineage.

Two analytic techniques were applied to morphological
character state distributions in order to estimate parasite
phylogenies: component analysis and Farris optimization.
Polarity sequences for individual characters were determined
on the basis of out-group comparison and ontogenetic argu—
ments. Data sources included museum specimens and litera-
ture descriptions.

Before analyzing the relationships of schistosome

genera, an out-group was designated in an analysis of

 

the

revez

Allan Christopher Carmichael

the families of the suborder Strigeata. This analysis
revealed the blood flukes of turtles, family Spirorchidae
as the out—group of the Schistosomatidae. Family level
analysis also called into question monophyly of the super—
family Schistosomatoidea by grouping the Clinostomatidae
and Sanguinicolidae together as a sister lineage of the
schistosomes and spirorchids. Analysis of host distri-
butions suggested that a three-host life cycle involving
pulmonate gastropods and vertebrate intermediate and defi-
nite hosts is a primitive pattern among the Strigeata.
Phylogenetic analysis of the Schistosomatidae revealed
three major clades: one composed of the genera assigned

to the subfamilies Bilharziellinae and Gigantobilharziinae,

 

a second branch including members of the genus Macrobil-
harzia and a. final clade involving all other members of
the Schistosomatinae.
Examination of host distributions revealed it pattern
of both host shifting and coevolution at both levels of
host involvement. Pulmonates are the primitive intermediate
host taxon and several shifts into prosobranch and opistho—
branch gastropods are indicated. While it appears that
birds primitively host the schistosomes, a minimum of
two independent shifts into mammals have occurred. Within
the Schistosomatinae, one shift is suggested by ancestors
of Schistosoma and Orientobilharzia, while at least one
other shift indicated is in the clade containing Bivitello-

bilharzia, Schistosomatium and Heterobilharzia. Biological

_

 

 

 

 

 

ACKNOWLEDGEMENTS

I wish to express my appreciation to the members
of my guidance committee, Drs. Donald 0. Straney, James
L. Edwards, John A. King and Jeffrey F. Williams for their
support and enthusiasm during this study. As committee
chairperson, Dr. Donald Straney provided invaluable training
and encouragement in developing this research.

My studies were partially supported by the Department
of Zoology and lnr a Doctoral Fellowship from the College
of Natural Science, Michigan State University.

My parents, Allan and Jean Carmichael have provided
continued support and encouragement throughout my graduate
career. Finally, I would like to thank Terry Stein for

his caring and support through all phases of this project.

iii

 

. :r:
m Flaws ,.

iv

12
14

18

33
33

36

47
47
48
74
84

96

 

 

 

THE SCHISTOSOMATIDAE .
Survey of Characters

Genus Level Characters

Phylogenetic Analysis.

Biological Considerations.

APPENDIX I

LIST OF REFERENCES

 

Page
tribution of Families in the
Suborder Strigeata . . . . 26
Farley's (1971) Classification of the
Schistosomatidae and Host Distribution. . . . . . . 30
Character State Distributions of the Strlgeata . 76
Components Spec1fied by Character State
istributions of the Strigeata. 77
Character State Distributions of the
chlstosomatidae. . 157
Components Spe01f1ed by Character State
lstributions of the Sohlstosomatidae 161

vi

 

LIST OF FIGURES

Page
1. Generalized life cycle of the strigeiform
trematode Diplostomum baeri (Diplostomatidae) . . . 19
2. Translation of data on relationships into
components and cladograms . . . . . . . . . . . . . 39
3. Farris Optimization . . . . . . . . . . . . . . . . 43
4. Furcocercous cercariae. . . . . . . . . . . . . . . 49
5. Position of cercarial excretory pore. . . . . . . . 52
6. Organization of cercarial anterior end. . . . . . . 59
7. Lophocerous cercaria (=Lophocercaria) . . . . . . . 61
8. Adult strigeiform trematode displaying the
tribocytic organ. . . . . . . . . . . . . . . . . . 65
9. Condition of intestinal cecae . . . . . . . . . . . 68
10. Esophageal gland cells present. . . . . . . . . . . 72
11. Component analysis of the Strigeata . . . . . . . . 79
12. Classical interpretation of relationships of
the superfamilies of the Strigeata. . . . . . . . . 86
13. Vertebrate host distributions assigned to the
preferred cladogram depicting relationships of
the Strigeata . . . . . . . . . . . . . . . . . . . 89
14. Intermediate invertebrate host taxa assigned
to the preferred cladogram dipicting relation-
ships of the Strigeata. . . . . . . . . . . . . . . 94
15. Shape of testes . . . . . . . . . . . . . . . . . . 100
16. Location of testes. . . . . . . . . . . . . . . . . 103
17. Dendritic branches present on paired cecae. . . . . 105

vii

 

4

viii
Page
18. Form of gynacophoric canal. . . . . . . . . . . . . 108
19. Ovary shape . . . . . . . . . . . . . . . . . . . . 113
20. Arrangement of vitelline follicles. . . . . . . . . 118
21. Presence of dendritic branches on the common
cecum of Dendritobilharzia. . . . . . . . . . . . . 124
22. Posotion of testes and genital pore . . . . . . . . 127
23. Position of female genital pore in Dendrito-
bilharzia and Gigantobilharzia. . . . . . . . . . . 130
24. Cirrus surrounded by prostate cells . . . . . . . . 135
25. Size and shape of seminal receptacle. . . . . . . . 141
26. Spatulate projections of posterior extremity
of Trichobilharzia. . . . . . . . . . . . . . . . . 144
27. Location of male genital pore . . . . . . . . . . . 148
28. Length of common cecum. . . . . . . . . . . . . . . 151
29. Component analysis of the Schistosomatidae. . . . . 166
30. Preferred cladogram depicting the phylogenetic
relationships of the Schistosomatidae . . . . . . . 186
31. Vertebrate host distributions assigned to the
preferred cladogram depicting relationships of
the Schistosomatidae. . . . . . . . . . . . . . . . 189
32. Intermediate host distributions assigned to
the preferred cladogram depicting relationships
of the Schistosomatidae . . . . . . . . . . . . . . 194
33. Probable phylogenetic relationships of the
molluscan intermediate hosts of the Schisto—
somatidae . . . . . . . . . . . . . . . . . . . . . 197
34. Farris optimization of schistosome geographic
distributions . . . . . . . . . . . . . . . . . . . 200
35. Proposed sequence of relationships between land
masses during continental breakup . . . . . . . . . 202

 

INTRODUCTION

Due to the nature and complexity of their life cycle,
parasites have long evoked fascination from biological
researchers and the general public alike. The close asso-
ciation between parasites and their hosts has intrigued
evolutionary biologists for decades, but in the absence
of a fossil record, biologists have done little more than
speculate on the origin and evolution of the host-parasite
interaction. The speciation mechanisms of parasitic organ-
isms have received some attention (Price, 1980), but speci—
ation studies have been constrained by a lack of knowledge
concerning species boundaries and the common lg priori
assumption that parasite speciation is dependent upon
prior host speciation. One area. which, until recently,
has been almost entirely neglected is that of parasite
Phylogeny.

This study was in part prompted by the development
of modern systematic techniques that I used to generate
parasite phylogenies independent of host bias. In addition
to the baseline work in censtructing 21 phylogeny for an
unexamined group 10f trematodes, I addressed several ques—
tions of general interest to evolutionary biologists.

My main concern was to evaluate what current distribution

 

of parasites within their host organisms reveals about
the pattern and process of parasite evolution. This ques-
tion can be approached by examining the degree of overlap
between independently derived phylogenies of parasites
and their hosts. Areas of congruence may suggest a history
of coevolution; discontinuities may indicate the occurrence
of host shifts during lineage formation. Parasite evolution
suggests a ready comparison with vicariance biogeography.
An obvious parallel exists between the distribution of
parasites in or on their speciating hosts and the geographic
distribution of organisms on moving land masses. Host
speciation can be likened to the formation of a geographic
barrier which isolates parasite populations, while host
shifts can be compared with organismal dispersal.

Before discussing these issues as well as the specific
applications of modern systematic methods to my study,
I will briefly review the history of evolutionary parasi—
tology, addressing implications for my current study.
I will also consider the problems inherent in conducting
research on parasitic organisms, problems which. in large
part explain why many evolutionary questions have not

been addressed to any great extent until recently.

Parasitism and Evolutionary Theory

Although definitions abound, parasitism generally
designates a symbiotic (sensu Cheng, 1970) association

Of two organisms of different species in which one of

 

the
char;
taxan

”helm

”.qu - - _| -

the two is metabolically dependent upon the other. This
characterization includes an array of organisms with diverse
taxanomic affinities: viruses, bacteria, protozoans,
"helminths," phytophagous insects, parasitoids, ectopara-
sitic arthropods, as well as numerous other taxa including
coelenterates, molluscs and vertebrates. Although parasites
taken as a whole do not form a natural group, Rensch (1959)
and other evolutionary biologists have noted similar morpho-
logical trends in the evolution of many parasitic taxa.
These include loss or reduction of limbs, sense organs,
and certain organ systems, as well as an increased repro-
ductive potential. Such similarities imply that related
processes may be involved in the evolution and maintenance
of parasitism in taxonomically distinct organisms.
Reflecting their common lifestyle, parasitic organisms
are frequently treated in a collective manner that assumes
they are governed by a unified set of biological principles.
Questions concerning the evolutionary processes involved
in deriving the morphological similarities that often
characterize parasites have yet to be addressed in an
empirical manner. For example, the environmental and
evolutionary forces impinging upon a gut dwelling helminth
and a phytophagous insect are almost certainly very differ-
ent, although both are considered ”parasites." Until
a better understanding of the mechanisnm; of evolutionary
change is available, I think it is wise to avoid g priori

generalizations which encompass all parasitic organisms.

 

 

For
platy

this

For this reason, my comments here are restricted to the
platyhelminth class Trematoda, which forms the basis of
this research.

With regard to theoretical considerations, most of
the focus of evolutionary parasitology has been directed
toward accounting for the acquisition of a parasitic life~
style by formerly free—living organisms. Faced ‘with the
lack of a fossil record which could be used to trace changes
in morphology associated with parasitism, researchers
have had to rely totally on extant organisms to aid in
the construction of their hypotheses. Where available,
comparisons with related free-living forms provide the
basis for most theories of the origin of parasitism. Impor—
tant contributions that consider the origins or platy—
helminth parasitism include those of Cheng (1973), Noble
and Noble (1976), Hyneman (1960), Llewellyn (1965), Janicki
(1920), Fuhrmann (1928), and Freeman (1973). These authors
have formulated elaborate scenarios to account for the
acquisition of a parasitic lifestyle.

Some evolutionary studies have focused on recent
parasite groups, but little empirical evidence has been
bought to bear on questions concerning the mechanisms
Of change (Price, 1980). The process of speciation among
parasitic organisms has received a moderate amount of
attention, although many of the results are anecdotal
(e.g., Inglis, 1971). Tabulations of the distributions

0f parasites and their host species have led to the

 

form
the

main
intim
The ‘

taken

formalization of three ”rules” which neatly summarize
the bulk of past thought on parasite speciation. The
main theme present in each of these rules is that of an
intimate association between host and parasite phylogeny.

The "parasitic rules" are presented below in a summary

taken from Noble and Noble (1976).

The Eichler Rule. When a large taxonomic
group (e.g., family) of hosts consisting
of many species is compared with an equivalent
group consisting of few representatives,
the large group has the greater diversity
of parasitic fauna.

The Szidat Rule. The more specialized the
host group, the more specialized are its
parasites. Hence, the degree of specialization
may serve as a clue to the relative phylo-
genetic ages of the hosts.

The Fahrenholz Rule. Common ancestors of
present~day parasites were themselves parasites
of the common ancestors of present-day hosts.
Degrees of relationship between modern para-
sites thus provide clues as to the parentage
of modern hosts.

In every presentation of these "rules," the fact
that numerous exceptions occur is mentioned (e.g., Noble
and Noble, 1976). Relatively little attention has been
paid to these exceptions, which may in fact outnumber
the cases that follow the ”rules." The concept of associa-
tion by descent has been firmly entrenched since the early
part of this century (reviewed in Stunkard, 1957), but
few studies have objectively examined the relationship
between host and parasite phylogenies. The views inherent
in the parasitic "rules" have conceptually limited studies

Of the pattern and process of parasite evolution, since

 

 

an a
paras
the p

preta

is $81

Specie

 

an I; priori assumption of congruence between host and
parasite phylogenies underlies each rule. Far too often
the pattern observed in nature is forced to fit an inter-
pretation of association by descent (e.g., Stunkard, 1970).
The most frequent manifestation of host biased taxonomy
is seen in the naming of new parasite species. Multiple
species of parasites are often designated when similar
forms are recovered from different host species. Descrip~
tions are frequently based on limited material without
addressing questions of intraspecific variation. Thus
inflated species reports result from the misinterpretation
of normal variation within species. Problems such as

these result from a lack of understanding of both individual

and populational variation in morphology.

M1211
Species designations in parasitic trematodes seldom
take into account the confounding influences of intra—
and interspecific variability. For example, although
many species distinctions are drawn on the basis of size,
the factors which influence size are seldom, if ever,
considered in a systematic context. Below II will review
the existing literature concerning individual and popu-
lational variation in trematodes and discuss the importance
0f variability in systematic studies.
Probably the single most important factor influencing

individual variability 511 parasites is the impact of the

 

host

ontog
envir
may h
impaci

defens

 

host organism (Wakelin, 1976). Host induced variation
is poorly understood, at least in part because parasite
ontogeny is difficult to observe independent of the host
environment. Although the nutritional status of the host
may have some effect on parasite morphology, host organisms
impact the development of parasites primarily via immune
defenses. The most frequently cited effects of immune
repsonse are stunted growth and diminished reproductive
output (Kennedy, 1975).

Host immune response may alter parasite morphology
by affecting rates of growth and sexual maturation.
Although in most cases the mechanisms of change remain
poorly understood, it is likely that both humoral and
cellular agents are involved (see Wakelin, 1976, for a
discussion and bibliography concerning host response).
The strength of the immune response differs between individ—
uals of the same host species and is dependent upon numerous
intrinsic and extrinsic factors. These factors include
age of host, state of immune responsiveness due to con-
current infections, intensity of parasitic challenge and
prior history of infection by the same or different species
of parasite (Kennedy, 1975). In addition, rates of growth
and development may be influenced by both inter- and intra-
Specific competition between parasites within a single
host individual, as noted by Kennedy (1975). Finally,
the nature and intensity of immune response varies between

host species, providing different opportunities and

 

chall
poter
both

growt
decre

repro<

 

challenges to invading parasitic organism. Thus the
potential for variability in parasite morphology differs
both within and between host species. Alterations in
growth rate and reproductive output are reflected in a
decrease in overall body size and the failure of
reproductive organs to reach full development. Helminth
species have historically been designated on the basis
of minor morphological differences that are often. metric
in nature, but very few studies consider the effects of
host response on the characteristics used in species
descriptions.

Studies which document the morphological variability
peculiar to a specific host-parasite system across its
range of immunologic potentials are, to my knowledge,
unavailable. Somewhat better data are available which
chronicle the variability induced by different host species
upon a single species of parasite, but the immunologic
and/or competitive factors which result in differing morpho-
logies have not been examined in most of these studies.
For example, Haley (1962) examined some of the complexities
Of intraspecific variability and reviewed earlier work
by Schiller (1959a, b, c) on morphological variability
in hymenolepid tapeworms. In addition, several authors
have examined intraspecific variability and its impact
upon systematic studies of particular trematode groups.

In a most comprehensive study, Blankespoor (1974) examined

morphological variability in Plagiorchis noblei (Trematoda:

 

 

Plagiorchiidae) which occurs naturally in the red-winged
blackbird, Agelaius phoeniceus. He infected 51 species
of birds and mammals with metacercariae reared in the
laboratory and collected fifteen hundred adult worms from
the 17 species of birds and mammals that were suitable
hosts. The following previously used characters were
unreliable in differentiating species within the genus:
body size, size of suckers, position of oral sucker, length
of esophagus, extent of vitellaria, size and position
of gonads, and size and position of cirrus sac. The degree
of variability in size or position in each character in
Plagiorchis noblei overlapped the ranges given in descrip~
tions of other species. Blankespoor did not comment on
the proximal causes of the observed variability but he
did note that Plagiorchis noblei failed to mature sexually
in several host species. He concluded that the only stable
adult characters which had been previously used in differen-
tiating species of the group were the size ratio of suckers
and intra-uterine egg size.

In a similar but less extensive study, Kinsella (1971)
evaluated host-induced variation in Quingueserialis guingue—
serialis (Trematoda: Notocotylidae). On the basis of
the variability revealed in his study, Kinsella revised
the genus, reducing the number of species from five to
three. Among the characters influenced by development
in different host species were the pattern of uterine

folding and certain body and sucker size ratios. Once

 

 

 

 

10

again, both metric and positional characters which had
previously been used in species descriptions proved unreli-
able due to extensive host induced variation.

Additional smaller scale studies address size variation
in whole worms or their constituent organ systems. Bruce
et al. (1961) found significant differences in the size
of reproductive organs from Schistosoma mansoni maturing
in twelve mammalian species exposed to infection. Differ—
ences in the degree of genital development (size of ovary,
testes, etc.) were specifically examined by Berrie (1960),
who raised Diplostomum phoxini, normally an avian parasite,
in both birds and mice. Works by MacKenzie and McKenzie
(1980), Pojmanska (1967) and Watertor (1967) focused primar-
ily on host-induced size differences and their relationships
to species level taxonomy of particular genera.

All of these studies examined either qualitative
or quantitative characters routinely used in the differen—
tiation of helminth species. Each study revealed levels
of host—induced morphological variability which exceed
the ranges commonly accepted when distinguishing species.
These studies demonstrate clearly that until more infor-
mation is available, characters which are defined by size
differences should be avoided in phylogenetic analysis
of trematode taxa.

Another aspect of variability that has particular
importance in systematic studies is ontogenetic variability.

In order to discern morpological changes due to an

 

alte:
be Cl
host

for v

 

alteration of normal development, ontogenetic series must
be compared using parasites raised in a number of different
host individuals and/or species. Such series are available
for very few trematode species.

The absence of adequate developmental information
has led to the description of new ”species" which are,
in fact, different development stages of a single species.
Goodchild and Martin (1969) provided a good example of
the difficulties in discerning ontogenetic synonomy and
the potential impact errors in the interpretation of
development can have on trematode systematics. In their
study of the trematode genus Spirorchis (Spirorchidae),
the authors synonomized three previously described ”species”
as ontogenetic stages of the chelonian blood fluke Spiror—
ghig scripta. The position of certain testes relative
to the cecal bifurcation changes during the development
of _§. scripta and the observed variation overlapped the
distinguishing features of the two other species.

One further level of variability seldom considered
in the literature on parasitic helminths is that of intra—
specific taxonomic variability. The concept of subspecies
is virtually non—existent in parasite taxonomy. The geo-
graphic sampling of a species which is required to designate
separate races is absent for all but a few helminths of
medical importance. Where studies do exist, the characters
examined are usually differences in biochemical markers

0r in ability to infect certain hosts, rather than strict

 

 

 

 

12

morphological variation (see Taylor and Muller, 1976,
and Taylor, 1965, for examples). In one of the few works
which did consider morphology, Hsu and Hsu (1957) examined
geographic variation in testes arrangement of Schistosoma
japonicum. Depending on the locality, testes were either
clustered or arranged in tandem. This level of variablity
would have been used to define separate species of a less
well understood organism. The importance of biochemical
variation has not been integrated into the systematics
of parasitic helminths, but the recent proliferation of
such work indicates that information valuable to systematic
studies will come from biochemical sources in the future

(e.g., Ross et al., 1978).

Phylogentic Analysis

The problems of host induced and/or geographic varia-
bility have serious implications for evolutionary studies
concerning host—parasite coevolution. Hypotheses of evolu—
tionary process cannot be reasonably advanced when the
taxa involved are inadequately known. At the species
level, problems of variability may be addressed through
detailed morphological studies that include data on both
ontogenetic and host induced variation. If such detailed
studies are unavailable, it may be more productive to
conduct an analysis of parasite relationships at a higher
taxonomic level which displays a more consistent pattern

of variation. Genus and family level characters are less

 

 

 

13

likely to involve size differences of the sort shown to
vary in the studies I have mentioned. In any case, the
first step in any study of host-parasite coevolution should
involve careful construction of a parasite phylogeny that
takes into account the problems of morphological variation.

The relatively recent rise of phylogenetic systematics
has provided improved methods of inquiry aimed at discerning
the patterns of relationships among organisms. The concepts
of phylogenetic or cladistic analysis were first articulated
by the German entomologist Willi Hennig during the early
19505, but did not have a major impact upon systematic
thought until translated into English (Hennig, 1966).
The theories advanced by Hennig have been greatly developed
and expanded during the past decade. A detailed discussion
of current cladistic theory and nethodology is available
in a number of recent texts including Wiley (1981), Eldredge
and Cracraft (1980), and Nelson and Platnick (1981).

In the present study, I chose cladistic analysis
as the best method of producing the parasite phylogeny
on which I based my inquiry into host—parasite coevolution.
The strength of cladistic methodology lies in 21 detailed
analysis of the patterns among the various characters,
in this case, morphological attributes of the organisms
under study. The distribution of each character or char—
acter state across taxa constitutes an individual evolu—
tionary statement. It is the careful knitting together

Of these individual statements across characters upon

 

 

14

which a phylogeny is based. Most of the arguments in
recent cladistic literature (for examples, see Funk and
Books, 1981; and Platnick and Funk, 1983) related to how
separate evolutionary statements are best integrated to
form phylogenies. I chose one particular method, component
analysis of Nelson and Platnick (1981), for use in the
present study because it allows all possible configurations
of the data to be examined; thus it does not limit g priori

the statements that can be made concerning a given phylo-

geny.

Objectives of the Study

The unique nature of the host—parasite relationship
presents a myriad of untapped opportunities for studies
of the coevolutionary process (Mitter and Brooks, 1983).
My research was prompted by the nearly total lack of modern
phylogenetic methodology in studies of helminth systematics.
Utilizing the techniques of phylogenetic analysis, I con—
structed a phylogeny for a particular group of parasites
in order 'U) examine hypotheses concerning the occurrence
of coevolution. Areas of congruence between host and
parasite phylogenies suggested a history of coevolution,
While host-shifting was indicated where branching patterns
were incongruent.

The group I examined was the trematode family Schisto—
somatidae. Because of the medical importance of some

0f its members, the family has been extensively studied

 

 

(Warr
a we]
world
pattel
taxa :

0f ho:

the v

source:

 

15

(Warren, 1973) and it is clear that its members form
a well—defined group. These blood flukes are distributed
world wide and are found in both birds and mammals. This
pattern of distribution in two distantly related host
taxa suggested a basis for examining competing hypotheses
of host-shifting and coevolution. Also, in contrast with
the vast majority of parasitic trematodes, sufficient
sources of data were available for use in this study in
the form of museum specimens and literature descriptions.

With the possible exception of the human blood flukes
Schistosoma mansoni and S. japonicum, most of the problems
in species designation discussed previously pertain within
the family Schistosomatidae. In addition, many species
and a few genera have been described on the basis of a
very limited sample, often consisting of one or a few
worms from a single host. For these reasons, I decided
to perform my analysis at the generic level, since most
genera are morphologically distinct and form discrete
assemblages. In cases where there were questions concerning
the congeneric status of certain species, I examined these
species separately.

By choosing to perform my analysis at the generic
level, I partially avoided the difficulties presented
by host-induced variation because schistosome genera show
clear morphological boundaries. To further minimize con—
founding problems of variation, in choosing characters

I focused upon major morphological structures which varied

 

betwe
Even

of ce
clear
varia‘

such 1

16

between genera but which were consistent within each genus.
Even in doing so, inconsistencies appeared in descriptions
of certain characters in the literature. It is not always
clear whether or not these inconsistencies represent natural
variation or inaccurate observation. I have addressed
such problems individually as they appear in my description

of characters.

In evaluating the relationships of genera within
the family Schistosomatidae, it was necessary to have
some understanding of the systematic status of related
taxa to perform an out-group analysis. The schistosomes
have traditionally been grouped with the other two families
of blood flukes from vertebrates, the Sanguinicolidae
of fishes and the Spirorchidae of turtles. Since there
are no detailed systematic studies examining the relation—
ships between the families of blood flukes, I first per—
formed family—level phylogenetic analysis in order to
determine the out-group of the Schistosomatidae.

Having a phylogeny I addressed the following questions
about schistosome evolution:

1. To what degree are host and parasite phylogenies con-
gruent? In order to answer this question, I compared
the phylogenies of parasite taxa and their definitive
hosts on the basis of published reports. In part
due to an extensive fossil record, vertebrate phylogeny
is one of the most robust phylogenies in the literature.

Although relatively little information is available

 

S(

W1

ve
th
co
pa
th:

isr.

 

17

with regard to intermediate host phylogeny, broad
patterns of molluscan host distribution allowed for
some degree of comparison.

What was the relative importance of host-shifting
versus coevolution in the phylogenetic history of
the Schistosomatidae? Coevolution is suggested by
congruence between host and parasite cladograms, while
patterns other than strict co—occurrence may indicate
that host—shifting occurred at some point in an organ-
ism’s lineage.

What is the relationship between phylogenetic analysis
of host-parasite distributions and historical bio-
geography? In the recent literature, there has been
a strong association between cladistic analyses and
historical biogeography (for a summary, see Cracraft,
1983). The main emphasis has been on the principles
of vicariance biogeography which examine patterns
of organismal distribution under the hypothesis that
observed taxonomic boundaries arose through geographic
isolation, for example, as a result of continental
drift. Vicariance biogeography is based on a model
of allopatric speciation and has a parallel in discus-
sions of host-parasite coevolution. Under such a
model, parasite speciation is dictated by prior host
speciation, just as host speciation is dictated by
the isolating events of changing land masses. There-

fore, I examined the extent to which parasite

 

 

 

distributions reflect Vicariant isolating effects
of host speciation as well as changes in geographic

distribution as a result of continental drift.

Schistosomatidae: Natural and Systemtic History

Blood flukes of the family Schistosomatidae belong
to the platyhelminth class Trematoda, subclass Digenea,
most of which are completely endoparasitic in vertebrates.
Digenetic trematodes are characterized by a complex life
cycle involving the production of several different larval
stages which develop in one or more intermediate hosts
(Figure l). Asexual reproduction frequently occurs in
several larval stages.

In all digenetic trematodes, a miracidium develops
in the trematode egg and, upon hatching, constitutes the
first free-living stage. Miracidia of most species are
characterized by a body covered with ciliated epidermal
plates (Figure l); the number and arrangement of the epi-
dermal plates have frequently been used in determining
relationships among trematode taxa. The miracidium is
either eaten In; or penetrates an invertebrate, most fre-
quently a gastropod mollusc. Within the mollusc serving
as an intermediate host, the miracidium undergoes metamor-
phosis to the next larval stage, which may be either a
SDOrocyst or redia. Rediae are characterized by the pre-
sence of a pharynx and intestine, which are absent in

SPorocysts. Sporocysts produce either another generation

 

 

 

 

 

 

Figure l.

19

Generalized life cycle of the strigeiform trema-

tode Diplostomum baeri (Diplostomatidae) (adapted

from Olsen, 1974).

Adult

Operculate egg

Miracidium

Redia

A.
B
C
D. Sporocyst
E
F Cercaria
G

Metacercaria

/\/®/?

2()

_

a
m
e
r.

 

.apted

of s;
rediae
ing up
produc
leaves

encyst:

 

H
21

of sporocysts (daughter sporocysts) or a generation of
rediae. Two generations of rediae may also occur. Depend—
ing upon the trematode species, either sporocysts or rediae
produce cercariae. The cercarial stage of most trematodes
leaves the mollusc and after at brief free-living period
encysts in or on either vegetation or a second intermediate
host, or penetrates the definitive host directly. In
cases where encystement occurs, the infective stage is
called a metacercaria. The metacercaria transforms into
a sexually mature adult when ingested by a definitive
vertebrate host. The second intermediate host is eliminated
in trematodes which penetrate their final host directly.
Above the generic level, much of trematode systematics
rests upon a framework based on larval characteristics.
LaRue (1957) utilized larval morphology to construct the
first substantive higher taxonomy of the digenetic trema—
todes. On the basis of detailed developmental studies,
LaRue divided the trematodes into two major groups according
to patterns of cercarial excretory bladder formation.
Within the group retaining an epithelial bladder, LaRue
erected the suborder Strigeata (order Strigeatida) to
contain the Clinostomatidae, the blood flukes (Sanguini-
colidae, Spirorchidae and Schistosomatidae) and the
so-called strigeiform families (Strigeidae, Diplostomatidae,
Proterodiplostomatidae, Cyathocotylidae, etc.). Families
Within the Strigeata possess a distinctive cercarial morpho—

logy in which the tail terminates in a prominent fork.

 

The

givil

ordin

my 3.1]

 

 

22

The terminal branches of the tail are termed furcae, thus
giving rise to the name furcocercous cercariae.

In the present study, I have accepted LaRue's sub-
ordinal classification as a basis upon which to conduct
my analysis. In a study which was relatively sophisticated
for the time in which it was conducted, LaRue examined
consistent patterns of larval variation and utilized
detailed developmental information in constructing his
taxonomy. LaRue avoided simplistic analyses of adult
morphology, the stage in ‘which problems of homoplasy and
host induced variation appear to be most prominent. Most
importantly, LaRue's higher classifiction of the digenetic
trematodes is, for the most part, not based on host rela-
tionships; thus host bias at least at the ordinal level
appears minimal. Because of these aspects of LaRue's
study, I felt his taxonomic framework provided a reasonable
place to begin a search for the outgroup of the Schisto-
somatidae.

The families within the suborder Strigeata are usually
divided into three superfamilies (LaRue, 1957). These
are the Strigeoidea containing the strigeiform trematodes,
the Clinostomatoidea, consisting of the Clinostomatidae
and the Schistosomatoidea, composed of the three blood
fluke families (Sanguinicolidae, Spirorchidae and Schisto-
somatidae). The relationships between superfamilies have
been discussed only briefly in the literature. Without

supplying data in support of his hypothesis, Short (1983)

 

sugge

are 1
fluke
histo
clino

the b

suggested that the strigeiform families and clinostomes
are more closely related to each other than to the blood
flukes. In a recent phylogenetic analysis, based on life
history stages, Brooks et a1. (1983) proposed that the
clinostomes are primitive relative to a clade containing
the blood flukes and strigeiform families.

Largely (M1 the basis of their blood inhabiting life-
style, the schistosomes have traditionally been grouped
with the other two families of vertebrate blood flukes.
Ohdner (1912) was the first worker to suggest that the
blood flukes from endo- and exothermic vertebrates were
related; Ward (1921) also voiced this opinion. In addition
to host environment, certain morphological similarities
between the families were noted, consisting of the absence
of a pharnyx and apparent loss of the redial stage.

Life cycle features are actually poor indicators
of relationship among the blood fluke families because
the strigeiform trematodes also lack rediae and thus the
character is not unique to blood flukes. It is useful
to note that this is the sole character upon which Brooks
et a1. (1983) unite the Schistosomatoidea and Strigeoidea.

Early attempts to clarify relationships between the
blood fluke families were rooted in the evolutionary con-
cepts and methodologies of the day. Reflecting a. trend
that was popular among evolutionary theorists until fairly
recently, a number of evolutionary ”series" were proposed

using the blood fluke families, in which various recent

 

 

 

form

form:

genus
from

homeot

 

forms were thought to have given rise to other recent
forms. At a time when very little information on any
group of trematodes was available, Ohdner (1912) proposed

the evolutionary series Liolope-Haplotrema-Bilharziella~

Ornithobilharzia-Schistosoma. In this series, the Spiror-

chid Haplotrema from turtles formed an important connecting
genus between the nonblood fluke Liolope (Liolopidae)
from amphibians and the more advanced schistosomes of
homeotherms. The liolopids are no longer considered to
be closely related to the blood flukes, and it is also
interesting to note that the blood flukes of humans are
considered to be derived from bird-inhabiting ancestors
in this series. In addition, the sanguinicolids do not
figure at all into Ohdner's scheme.

In a summary of his earlier papers, Mehra (1950)
suggested that the spirorchid subfamily Haplotrematinae
formed the primitive stock which gave rise to the Sanguini-
colidae along one line and the Schistosomatidae along
another. In proposing this sequence, Mehra differed from
the majority opinion expressed most clearly by Stunkard
(1921, 1923, 1970), who saw a logical derivation of spiror-
chids from sanguinicolids, and of schistosomes from spiror-
chids. This progressive series was quite compatible xvith
interpretations of host evolution and Stunkard went as
far as to suggest that the blood flukes were an ancient

group, antedating the origin of birds and mammals.

 

 

 

bili
stag.
Stril
in T:
inteI

for

 

25

Families within the Strigeata show considerable varia-
bility with regard to life cycle pattern. Host and larval
stage distributions for the families of the suborder
Strigeata (n1 which I focus 111 my analysis are presented
in Table 1. All of these families possess an invertebrate
intermetiate host and a vertebrate definitive host. Except
for some questionable reports of sanguinicolid cercariae
developing in pelecypods and annelid worms, invertebrate
hosts are always gastropod molluscs. In addition, an
intermediate vertebrate host, and thus a metacercariae
stage, occurs in all families other than the three blood
flukes. Fish usually serve as the intermediate vertebrate
hosts, although amphibians (tadpoles) may serve this role
for a few species in the families Strigeidae and IJiplo—
stomatidae (Dubois, 1938). Finally, cercariae develop
from daughter sporocysts in all families except the clino—
stomes, where a redial stage is present.

Relative to the Schistosomatidae, comparatively little
is known about the biology of the other blood flukes.
Although sanguinicolids are found in a diversity of marine
and freshwater fishes, very few life cycles appear in
the literature. While relatively more information is
available concerning spirorchids inhabiting freshwater
turtles, virtually nothing is known about marine species.
Smith‘s (1972) review of the sanguincolids and spirorchids
remains the only major work published on either family

in recent years.

 

 

 

I'll. hill

0.” hh§<h .RC ZOHrHDmHmHmHQ .HWONN
H. mqmawrfi

26

 

 

menocmanonpwwmo
meEEmE necozmsnomosm
mcawm mpscoaasm oscfipwsowOpmHnom
assocmnnomOhm
mmfipnsa wpncosasm ocpwaon0pflqm
wmcomaooaom
wmshos beacons
wflnocsanomoam
zmfim mpdcosasm ospflaoosnflsmswm
wzdfiHHoOoOpo
mcnflm swam dunnosasm odepaEOmeCHHo
massed:
mchfim
moaflpmmm swam mumsosasm mmcflazpooonpmao
wsdfiaficoooeo swam mpmsosasm owchsEOPmoHQHconouonm
massed: Amoaomcmev mqwflnflaos<
mcnwm nmfim newsoaasa owchwEOpmonflm
masses: Amoaochev mqaflnflnas<
mohwm swam «umsoaase ompHomepw
pmom pmom nmom sigma
opmpnopeo> oancoanpsH ouwficosbopuH
o>HpH2Hon oussnopeo> casebopeoecH
so awash

 

 

<E<m0HmBm mmflmombm mmfi ZH mmAHE<m m0 ZOHBDmHmbHQ Emom

H mqm¢8

 

Schis
due n«
of tk
group
specie
tremat

are se

 

27

In contrast with the other blood flukes, the family
Schistosomatidae has received considerable attention,
due no doubt to its human inhabiting members. Blood flukes
of the family Schistosomatidae comprise a welledefined
group in which all members are dioecius. Although dioecius
species and even genera occur in other familes of digenetic
trematodes, the schistosomes are unique in that the sexes
are separate in all species.

The first species of schistosome was recorded by
Rudolphi (1819) and was named Distoma canaliculatum; this
avian blood fluke is currently assigned to the genus Qgpi;
thobilharzia. The description of the first human schisto—
some, Distoma haematobium by Bilharz in 1852 was followed
by a long period of taxonomic confusion. The genus of
the human blood flukes was assigned various names including
Schistosoma (Weinland, 1858), Gynaecophorus (Diesing,
1858), Bilharzia (Cobbold, 1859) and Thecosoma (Moquin-
Tandon, 1860). The names Schistosoma and. Bilharzia were
used interchangeably until 1954, when the International
Commission on Zoological Nomenclature ruled the generic
name of the group Schistosoma, while the term ”bilharziasis”
should be used to designate the disease.

Stiles and Hassal (1898) assigned the schistosomes
to the subfamily Schistosominae of the family Fasciolidae.
Loss (1899) raised the group to family rank, the Schisto—
somidae, and Poche (1907) grammatically revised the family

name, replacing it with the term Schistosomatidae. A

 

 

mm

majo
1913

of S.

the :
24 s]
on th
The t
and u
alth01
Bilhal
lines

in how

 

28

major breakthrough in understanding this group came in
1913 when Miyairi and Suzuki demonstrated the life cycle
of i W.

Price (1929) provided the first major summary of
the family, which at that time contained nine genera and
24 species. He divided the family into two subfamilies
on the basis of relative length of the gynacophoric canal.
The two subfamlies erected by Price, the Schistosomatinae
and the Bilharziellinae, are still accepted by most authors,
although there have been frequent modifications of the
Bilharziellinae. This subfamilial split does not follow
lines of host affinity; the Schistosomatinae are found
in both birds and mammals, while the Bilharziellinae are
strictly avian.

Mehra (1940) raised the genera Gigantobilharzia and

Dendritobilharizia to subfamily rank: The Gigantobilhar-
ziinae and Dendritobilharziinae. Although this separation

was accepted by both Skrjabin (1951) and. Yamaguti (1971)
in their volumes on digenetic trematodes, there is a major
problem with Mehra's division. Although his rationale
is not entirely clear, the subfamily Gigantobilharziinae
was apparently distinguished due to Mehra’s incorrect
assumption that the cirrus pouch was absent in the genus.
The subfamily Dendritobilharziinae was recognized on the
basis of the extensive dendritic branches on the common
cecum. Farley (1971) pointed out that females in both

subfamilies possess a genital pore which opens at the

 

 

ext]

of
Farl

the

frmm

famil

 

 

29

extreme anterior end, next to the mouth. On the basis
of this arrangement which is unique within the family,
Farley grouped the two genera into a single subfamily,
the Gigantobilharziinae.

The only major deviation from this higher taxonomic
framework comes from Azimov (1970), who grouped the three
families of vertebrate blood flukes into a single order,
the Schistosomatida. The schistosomes were divided ”on
the basis of morphological and ecological peculiarities"
into two families: the Schistosomatidae of mammals and
the Ornithobilhariidae of birds. This host-based scheme
of classification has not been accepted widely and to
my knowledge has been used only by Smith (1972).

In a comprehensive manner, Farley (1971) summarized
the literature up to that time and gave a taxonomic history
of generic classification of the Schistosomatidae. He
also synthesized both generic and specific taxonomies
and listed known schistosome species. In reading Farley’s
summary, it is clear that although generic limits are
fairly well defined, there is much confusion with regard
to species designation. With two exceptions discussed
below under Methods, my analysis will utilize Farley’s
generic designations, presented in Table 2 along with
host distributions. Farley's work stands in sharp contrast
with that of Yamaguti (1971), who provided the summary
0f digenetic trematodes most widely used today. Since

little or no attention is paid to problems of intraspecific

 

 

 

30

TABLE 2

FARLEY'S (1971) CLASSIFICATION OF THE
SCHISTOSOMATIDAE AND HOST DISTRIBUTIONS

Subfamilies
and Genera

Schistosomatinae
Schistosomatium

Bivitellobilharzia

Heterobilharzia

Orientobilharzia

Schistosoma

Austrobilharzia

Macrobilharzia

Ornithobilharzia

Bilharziellinae
Bilharziella

Trichobilharzia

Intermediate
Hosts

Pulmonata
Lymnaeidae

Pulmonata
Lymnaeidae

Pulmonata
Lymnaeidae

Pulmonata
Lymnaeidae

Pulmonata
*Lymnaeidae

Planorbidae
*Prosobranchia

Prosobranchia

?

Prosobranchia

Pulmonata
Planorbidae

Pulmonata
Lymnaeidae
*Physidae

Definitive
Hosts

Rodentia

Proboscidea

Carnivora

Artiodactyla

Artiodactyla

Primates

Rodentia
*Carnivora
*Insectivora

Anseriformes
Charadriiformes

Pelecaniformes

Charadriiformes

Anseriformes
*Rallidae
*Charadriiformes
*Pelecaniformes
*Podicipediformes

Anseriformes
*Passeriformes

 

 

 

 

 

 

 

Table 2 (continued)

 

Subfamilies
and Genera

Gigantobilharziinae
Gigantobilharzia

Dendritobilharzia

31
Intermediate Definitive
Hosts Hosts
Pulmonata Anseriformes
Physidae *Passeriformes
Planobidae *Charadriiformes
*Opisthobranchia *Ardeiformes
*Podicipediformes
? Anseriformes
*Pelecaniformes

*Occurs in less than ten percent of the reports.

 

 

 

the
are ‘

the

 

METHODS

Selection of Taxonomic Samples

The first step in any phylogenetic study is determining
the taxonomic level at which the organisms in question
are to be analyzed. This is often not an easy task since
the validity of previously defined taxonomic boundaries
is often questionable, and sampling techniques of earlier
studies unclear. Carlton (1980) provided a useful discus-
sion of these problems in his analysis of the neotomine—
peromyscine rodents, where he utilized large population
samples of nominal species and adopted the exemplar method
of Sneath and Sokal (1973) in his characterization of
species and genera. In the present study, my attempt
to use Carlton's methodology met with a number of diffi—
culties.

With the exception of the human infecting schistosome
species, large population samples are unavailable for
blood flukes. All that is available for the majority
Of schistosomes are published descriptions and/or type
specimens. Specific and even generic descriptions are
frequently based on one or a few specimens and the majority
Of preserved specimens appear to reside in private collec-

tions that are not widely accessible.

33

 

the ‘
are t

the .

 

 

METHODS

Selection of Taxonomic Samples

The first step in any phylogenetic study is determining
the taxonomic level at which the organisms in question
are to be analyzed. This is often not an easy task since
the validity of previously defined taxonomic boundaries
is often questionable, and sampling techniques of earlier
studies unclear. Carlton (1980) provided a useful discus-
sion of these problems in his analysis of the neotomine—
peromyscine rodents, where he utilized large population
samples of nominal species and adopted the exemplar method
of Sneath and Sokal (1973) in his characterization of
species and genera. In the present study, my attempt
to use Carlton's methodology met with a number of diffi-
culties.

With the exception of the human infecting schistosome
species, large population samples are unavailable for
blood flukes. All that is available for the majority
Of schistosomes are published descriptions and/or type
specimens. Specific and even generic descriptions are
frequently based on one or a few specimens and the majority
0f preserved specimens appear to reside in private collec—

tions that are not widely accessible.

33

 

 

speci
furth
This,
of sp

used 1

my an

discus

 

34

The general lack of understanding concerning intra-
specific variability' and morphological species boundaries
further complicated my selection of taxa for analysis.
This, coupled. with an extremely host-biased proliferation
of species names makes the concept of "species" as currently
used in parasitology questionable at best.

To cope with these difficulties, I chose to perform
my analysis at the generic level. As II have previously
discussed, I feel that the schistosome genera represent
coherent systematic units. In nearly all cases, these
units can be unamibuously differentiated on the basis
of morphological characters. Where generic characters
are unclear, I performed separate analyses on conflicting
species or species groups. Analysis at the generic level
was facilitated because at least some specimens from the
majority of genera. were available from the U.S. National
Helminthological Collection which resides at the Animal
Parasitology Institute of the U.S.D.A. I used these speci-
mens to supplement and confirm the presence of morphological
characters mentioned in published descriptions. Thus
museum specimens and descriptions from the literature
were the major sources of morphological data in this study.

I limited my character analysis to relatively gross
morphological traits which varied between genera but were
constant within genera. I was forced to rely totally

upon published descriptions in determining the characters

 

of

and

w «1.311;;— . .

of certain poorly known genera (e.g., Bivitellobilharzia
and Afro-Asian Macrobilharzia).

In addition, I had. at my disposal numerous specimens
of Schistosomatium douthitti, Schistosoma mansoni and
§. japonicum that assisted greatly in interpreting material
prepared by other researchers. My choice of morphological
characters was limited by the nature of the available
material. Certain potentially useful characters which
Show consistent variation, such as tegumental spination,
require special preparation for clear observation and
therefore were not examined because I did not have access
to previously unstained material. Because they are best
observed on living specimens, larval characters were taken
entirely from published descriptions.

I chose to follow a modified version of Farley's
(1971) interpretation of the family Schistosomatidae.
Farley carefully synonymized previous descriptions and
grouped the schistosomes into a series of morphologically
discrete genera. I varied from his scheme in two instances
where character incongruities indicated that the proposed
genera were actually more complex than prior observations
suggested. The first exception is the genus Sinobilharzia,
a name used by Dutt and Srivastava (1961) to. designate
Ornithobilharzia odhneri (Faust, 1924). Although Farley
Placed this species in the genus Austrobilharzia” I feel
it is safer to regard it as a monotyic genus based on

features to be discussed in the survey of characters.

 

The
of
the
anal
01d
schi
the

info

soma‘
to b
anal}

geatz

 

_____———-——— 7 ,
M‘ (an. ’3‘

36

The second. exception concerns African and Indian reports
of the genus Mgppppilpppgig. Although I have retained
the Old World forms in the genus Aggagflgplpgpgig, Ii have
analyzed them separately due to inconsistencies between
Old and New World descriptions. The fourteen genera of
schistosomes examined are listed in Appendix I along with
the specimens and/or species descriptions used in assembling
information on each genus.

In order to analyze relationships within the Schisto-
somatidae, I needed to determine an appropriate family
to be used as zul out—group. To do this, I performed an
analysis of the eight families within the suborder Stri—
geata, as defined by LaRue (1957). The character states
and their distributions used in this analysis were taken
from reviews of the families given by Yamaguti (1971),
Dubois (1938), Smith (1972), Price, (1929), and Schell
(1970). In assessing family level character state trans—
formations, I utilized other famlies within the order

Strigeatida as out—groups of the Strigeata.

Character Descriptions and Analytic Methods

In contast with many systematic studies, there was
little previous work on comparative trematode morphology
available that I could use as a guide in choosing characters
for my analysis. Trematode taxonomy rests on an ill—defined
framework in which host relationships, morphology and

life history patterns have been knit together to form

 

 

 

 

wit
I 1
log
out
dett
use

and

 

37
a scheme of relationships. I found few useful characters
within this taxonomic framework. The characters that

I have defined for analysis are relatively gross morpho-
logical attributes, some of which display well defined
ontogenetic patterns. I have utilized two methods of
determining polarity sequences for multistate characters
used in the various analytic methods: out-group analysis
and ontogeny.

Out-group analysis is perhaps the most widely accepted
method used as an independent criterion for evaluating
evolutionary transformations (see Wiley, 1981, for complete
discussion). Character states shared by a taxon and its
out-group are assumed to have been inherited from a common
ancestor, and thus are termed. primitive (plesiomorphies).
Character states absent in the out-group but present in
some members of the taxon under study (apomorphies) may
indicate relationship between those members. The problems
in distinguishing homology and homoplasy are hopefully
minimized by using this method.

The use of ontogeny as an independent criterion for
establishing evolutionary transformations was formalized
by Nelson (1978). Recently Fink (1982) discussed the
conceptual relationship between ontogeny and phylogeny
relative to systematic studies and considered the problems
which heterochronic development posed for the evaluation

Of homology. Ontogenetic considerations of character

state changes involve a robust adaptation of the biogenetic

 

 

lav
of

abl
nor
cha
digl
rafl

ree‘

infel

Both

 

38

law in which it is assumed that apomorphic character states
of a clade go through ontogenetic stages that are recogniz—
able as plesiomorphic or embryonic states of that clade's
more primitive relatives, primitive with respect to the
character only. Since much of the higher taxonomy of
digenetic trematodes is based on development of cercariae,
rather detailed literature accounts were available for
reevaluation.

I have coded the qualitative characters that form
the basis of my analysis as nominal, discrete variables.
Most characters express either presence/absence (n? degree
of development of morphological structures and one character
(testes number) is meristic.

To generate hypotheses of phylogenetic relationship,
I utilized a combination of two methods of phylogenetic
inference: component analysis and Farris optimization.
Both are variations of the cladistic methodology proposed
by Hennig (1966) and each requires an 3 priori determination
of primitive and derived character polarities. Under
the component analysis method of Nelson and Platnick (1981),
the distribution of each character state across all taxa
defines a "component” of node of a hypothetical cladogram.
Figure 2, adapted from Straney (1980), illustrates how
character state distributions are translated into compo-

nents. These components are then knit together to form
branching diagrams. I utilized Felsenstein's largest

clique program to generate all possible combinations of

 

 

 

 

 

 

 

 

 

Figure 2.

39

Translation of data on relationships into compo-
nents and cladograms. The distribution of
three character states (1-3) across four taxa
(A—D) in (a) defines groups of taxa, or compo-
nents (b). These components can be combined
in a consistent manner to produce a cladogram
(0), although the set of disarticulated compo-
nents contains the same information as does
the full cladogram. The branching diagrams
assume that each taxon has been defined by

other unspecified character states (from Straney,

1980).

 

 

the

Arg
con'
natt
sta'

com]

 

41

the components defined by my data. This method generates
a large number of cladograms for a given set of data.
Arguing by analogy with experimental studies, cladograms
containing the most highly replicated components are desig-
nated as a smaller subset for detailed analysis of character
state distributions. For a more complete discussion of
component analysis, see Nelson and Platnick (1981).

The version of the largest clique program used in
this study was provided by Joseph Felsenstein of the Univer-
sity of Washington as part of his Phylogeny Inference
Package for use on an Apple II Plus computer. For docu—
mentation as well as further information on the assumptions
and methods of this technique, see Felsenstein (1978,
1979 and 1982).

After I chose a subset of cladograms that were corrobo—
rated because they included the most highly replicated
component or components, I utilized the principles of
Farris optimization to assign character state distributions.
Farris optimization was proposed by Farris (1970) and
formalized by Mickevich (1981) and is a parsimony method
that seeks the branching diagram that minimizes the total
number of character state transformations. Thus the clado—
gram involving the smallest number of steps is considered
the most reasonable estimate of cladistic history (see
Farris, 1983, for a, discussion. of parsimony). In. actual

fact, a number of minimal length cladograms may be gene-

rated. When I could not employ parsimony in choosing

 

 

 

 

bra
Hec
lem
def:
hert
ture
that
In

HOII‘

is t
tion
deri
limi

late

 

a preferred cladogram because the number of steps in each
branching diagram was equal, 1 adapted the arguments of
Hecht and Edwards (1977). Given two cladograms of equal
length, I chose the one with the smallest number of branches
defined by derived loss characters. The assumption. made
here is that two organisms are more likely to lose a struc-
ture in parallel (thus confounding the search for-homology)
than they are to evolve a new character state in parallel.
In other words, I favored cladograms defined by complex,
non-loss characters.

Fundamental to the workings of Farris optimization
is the fact that no restrictions on the direction of evolu—
tionary change are imposed, and therefore reversals from
derived back ‘to primitive states may occur. This greatly
limits the number of parallel events which must be postu-
lated to explain a given character distribution.

Under the optimization method, conflicting character
state distributions are assigned to intermediate nodes
Of a given cladogram. This technique minimizes the total
length of a cladogram, resolving multichotomies by combining
branches and thus reducing the number of parallel evolution-
ary changes.

Figure 3A portrays a cladogram consisting of seven
taxa (A—G), with the state distribution. of a. conflicting
character occurring in the branch containing taxa E, F
and. G. The hypothetical character state at an ancestral

node is unambiguous if all taxa derived from that ancestor

 

 

 

 

Figure 3. Farris Optimization.

 

44

   

I,O (union)

 

  
    

l (intersection)

1,0 (union)

 

 

 

 

 

 

pro
up;
tiv<
the
the
bran

chot

reVe

3F)

 

45

have the same state. This point is illustrated by nodes
a and b in Figure 3A, where the ancestral states 1. and
0, respectively, have been assigned. For the branch giving
rise to taxa E-G, the ancestral node is provisionally
assigned either the intersection of character states,
or the union of states when no intersection occurs (Figure
3B, C). If any ambiguities remain, as in nodes 0 and
e (Figure 3C), they are resolved by taking the intersection
of the ambiguous node and its immediate ancestor. This
process begins at the base of the cladogram and works
up; the most ancestral state is always assumed to be primi—
tive (Figure 3D). In Figure 3E, basal node e is assigned
the primitive state. At this point, branches containing
the derived state can be combined (Figure 3F). Optimized
branches are illustrated by dashed lines. Thus, multi—
chotomies are resolved and the number' of parallel events
is minimized. This procedure allows characters to undergo
reversal to the primitive state (as in taxon G, Figure
3F) and therefore does not force evolution in one direction.

A major goal of this study was to examine hypotheses
of cospeciation between parasites and their host taxa.
The first step in the process consisted of an independent
evaluation of the relationships between parasite genera
described above. The next step involved comparing the
Phylogenetic relationships of host taxa with those of
their parasite fauna, in a search for congruent branching

Patterns that would suggest a history of coevolution.

 

 

 

46

At the class level, the phylogenetic relationships between
vertebrate taxa are fairly well understood (McFarland
et al., 1979). Existing phylogenies provided a basis
for comparison with the family-level tree depicting the
relationships within the suborder Strigeata.

Likewise, existing host phylogenies were used in
assessing the phylogenetic congruence between schistosome
host groups. While reasonable phylogenies exist for mammals
(McKenna, 1975), the relationships among avian hosts as
well as among the pulmonates (and molluscs in general)
are far from clear. Although the modern approaches to
avian systematics by Cracraft (1981) and to molluscan
relationships by Hubendick (1978) are useful, I will address
the limitations of these incomplete phylogenies in my
discussion.

In my final analysis, 1 evaluated the preferred para—
site phylogeny in light of vicariance biogeographic theory,
which suggests that much of organismal evolution and distri-
bution reflects a pattern of isolation during continental
breakup. The concepts of vicariance biogeography have
been extensively developed by Brundin (1966), Humphries
(1981), Rosen (1978), and Nelson and Platnick (1981).
These authors provide cladograms depicting continental
breakup which I used in comparisons with organismal phylo-

genies generated in my analysis.

 

 

 

 

ger
use
are
of
in

8011]

THE STRIGEATA

Survey of Characters

My analysis and arguments concerning trematode phylo—
geny are presented in two sections. The 17 characters
used in analysis of families in the suborder Strigeata
are presented below followed by an analysis and discussion
of family level relationships. The 24 characters used
in my analysis of generic relationships within the Schisto—
somatidae begin on page 99. Each character is indexed
by a number that serves to identify it in the various
tables and figures. Character states are indicated by
numbers in parentheses, the inferred ancestral (plesio-
morphic) condition being (0) and derived (apomorphic)
states consisting of positive numbers. Character state
transformations usually follow numerical order, except
in instances of furcating character state trees; in such
cases, the immediately ancestral state is indicated next
to the character state description. The majority of charac—
ters are two state variables, but some contain up to five
character expressions. I have included discussions to
clarify the nature of character variations and problems
of homology, and to outline my rationale for determining

character polarities. In some cases, this treatment will

47

 

 

 

 

cond

to 1;]

 

48

be extensive since argumentation of this sort is lacking
in discussions of trematode phylogeny. As previously
mentioned, all polarities have been determined on the

basis of out-group comparison and ontogenetic arguments.

Family Level Characters

Character 1: Cercarial Body Form.
(0) Simple, straight tailed or tailless.
(l) Furcocercous, forked tails.

Out of the wide variety of cercarial body forms
observed in nature (see Schell, 1970, for examples), members
of the families in the suborder Strigeata all display
a similar developmental pattern leading to a forked tailed,
or furcocercous body form (Figure 4). The forked tailed
condition is presumably derived from a primitive straight
tailed or tailless ancestor. Ontogenetic studies by Hussey
(1941), Cort (1917), and Kuntz (1950) indicate that develop-
ment proceeds from tailless through straight tailed to
a final forked tailed form. This character is unique
to the order Strigeatida. The above—mentioned developmental
studies are used as justification for Characters 2 and

3 as well.

Character 2: Length of Furcae.
(0) Furcae short: Brevifurcate

(l) Furcae long: Longifurcate

 

 

 

 

 

 

 

 

Figure 4. Furcocercous cercariae.

A. Brevifurcate

B. Longifurcate

 

 

furcae,

Position of Cercarial Excretory Pore

(0) Terminal.
(1) Subterminal.

(2) Secondarily terminal.

Developmental studies show that cercariae possessing

subterminal excretory pores

(Figure 5B) pass through a
stage

This situation applies

for all reports of development within cercariae of the

Strigeata, with the exception of so called ”

vivax” cercariae

belonging to the

family Cyathocotylidae. Cercariae of

this poorly understood family possess terminal excretory

pores and are longifurcate. Although developmental studies

(Anderson and Cable, 1950'

secondarily derived from the subterminal

seen in other longifurcate cercariae. This

 

 

 

 

 

 

Figure 5.

Position of cercarial excretory pore.

A. Terminal

B. Subterminal

 

01
Th
me

no

Ch:

pre
The
is

in

 

 

54

of Linstowiella szidati by Anderson and Cable (1950).
The pores begin at the tips of the furcae in early develop-
ment, pass through :1 distinct subterminal phase and then

move once again to the tips of the greatly elongated furcae.

Character 4. Site of Cercarial Development.
(0) Redia.
(l) Sporocysts.

Cercariae develop in a number of different larval
precursors depending upon the family under examination.
The most widespread pattern among trematodes as a whole
is the production of cercariae from rediae which have
in turn been produced from sporocysts. Within the Strigeata
I consider the absence of the redial stage and subsequent
production of cercariae from sporocysts as a derived loss
condition because redia are widespread in the other families
of the order Strigeatida. Among the families involved
in this study, only the Clinostomatidae unambiguously
possess a redial stage in their life cycle. Although
all of the other families have been characterized by the
lack of redial stage, there are inconsistencies in reports
of the family Sanguinicolidae. Rediae are distinguished
from sporocysts because they possess a musculai' pharynx,
which sporocysts lack (Figure 1). There are four reports
of cercariae assigned to the family Sanguinicolidae that
develop in rediae. All four reports are problematic,

and only in the case of Cardicola (= Sanguinicola) davisi

 

 

 

alt
are
Cer
in

tre
pos
col
cla
sta;
stor
"sp<
and

frOn

Once

Cari

 

55

(Wales, 1958) has a complete life cycle been documented,
although the published report is sketchy and many details
are absent. McCoy (1929) and Croft (1933) reported that
Cercaria brevifurca and Q. whitentoni, respectively develop
in rediae. ("Cercaria" is the generic name assigned to
trematode cercariae of unknown affinities.) These cercariae
possess a dorsal-median fin fold, common to both sanguini-
colids and clinostomatids. Both cercariae appear to be
classic sanguinicolid lophocercaria larvae, but adult
stages which would distinguish these cercariae from clino-
stomatid larvae have not been described for these cercarial
”species." Likewise, Cercaria hartmanae (Martin, 1952)
and Q. amphicteis (Oglesby, 1961), which develop in redia
from annelid worms, are assumed to be sanguinicolids.
Once again, the adult stages are undescribed and the cer-
cariae possess certain morphological abnormalities which
I will discuss in a subsequent section.

These three examples highlight the fact that the
family Sanguinicolidae is poorly known and is, most likely,
highly variable. In order to address the potential varia-
bility seen III the Sanguincolidae, I coded this character
in two different manners, resulting in two different data
sets which I analyzed separately. First, until more sub-
stantial instances of cercarial development in redia. are
reported, I scored the sanguinicolids for the derived
state in which the redia is absent, as indicated. by the

majority of studies (see Smith, 1972, for an extensive

 

 

bi‘
thl
de‘
the
(1m
hay
nor

the

Cha

56

bibliography). Secondly, I coded the sanguinicolids for
the primitive state based on the limited reports of redial
development because the primitive state occurs within
the range of variation displayed by the taxon and the
derived state shown by certain members of the family may
have been independently achieved one or more times. Until
more data are available, I cannot distinguish between

the two alternative codings.

Character 5. Presence/Absence of Muscular Pharynx.
(O) Pharyngeate.
(l) Apharyngeate.

A muscular pharynx is found in both the cercariae
and adults of most digenetic trematodes. It is absent
and considered to be a derived loss character only in
adults and cercariae of the schistosomes and spirorchids.
The loss of ea pharynx has assumedly occurred in parallel
in adults of one genus (Apharyngostrigea) in the family
Strigeidae, since the pharynx is present in cercariae
(Oliver, 1940). There is one report of a pharynx present
in the cercariae of the sanguinicolid Cardicola davisi
(Wales, 1958). This is consistent with the reports of
redial development in the same study, and although unique,
there is currently no way to confirm or deny its validity.
Accordingly, I will provide two codings of this character
in two separate data sets, following the argument used

in considering Character 4. It is useful to note that

 

 

57

all pharyngeate lophocercariae (cercariae of the families
Sanguinicolidae and Clinostomatidae that possess a dorsal-
median fin fold) for which the‘ adults are unknown are
automatically considered to be clinostomatids. This may
underestimate the occurrence of pharynges in the sanguini-
colids. To further confuse matters, Agarwal (1959) notes
that cercariae of the clinostomatid genus Clinostomum
may lack a pharynx in both cercariae and adults. I assume
that this is a ‘unique loss event, since the Inajority of
clinostomes possess pharynges.

Loss of the pharynx along with the absence of a redial
stage were used to establish the superfamily Schistosoma—

toidea in the taxonomy of LaRue (1957).

Character 6. Miracidial Plate Pattern: Numbers of Plates
in the Four Consecutive Tiers.
(O) 6,8,4,3.
(1) 6,6,4,2.

The miracidia of most trematodes are covered with
tiers of ciliated plates. In nearly all families of the
order Strigeatida, the number of plates in each of the
four tiers is remarkably consistent: 6 in the first,
8 in the second, 4 in the third, and 3 in the fourth.
Within the Spirorchidae, a presumably autapomorphic condi-
tion occurs in which the number of plates in rows two

and four differ: 6.6.4.2.

 

 

 

6A

In
the

has

the

Ope

the

Cha:

 

 

 

openings. The head organ else among

the trematodes, and I assume it to be a derived, homologous

character state.

Character 8. Penetration Glands.

(0) One morphological type.

(1) Two morphological types.

The different types

of penetration glands found in

trematode cercariae are distinguished by their differential

staining ability.

take acidic stains. Most cercariae

possess penetration glands of a single morphological (stain-

ing) type. In schistosomes and spirorchids, the glands

are differentiated into an anterior mucoid group and a

Posterior lytic group. I will presume that the presence

 

 

 

 

 

 

59

Figure 6. Organization of cercarial anterior end.
A. Oral sucker present, mouth terminal

B. Head-organ present, mouth subterminal

 

 

 

 

 

 

 

 

 

Figure 7. Lophocerous cercaria (=Lophocercaria).

 

 

 

 

charac
Collec

known

shared

63

of these two gland types represents a homologous and derived
condition, because this arrangement is found nowhere else

among the Strigeatida.

Character 9: Cercarial Dorsal-Median Fin Fold.
(0) Absent.
(1) Present.

A prominant dorsal-median fin fold (Figure 7) is
characteristic of sanguinicolid and clinostomatid cercariae.
Collectively, cercariae possessing this character are
known as lophocercariae. There are some reports of presumed
sanguinicolid cercariae in which the fin fold may be absent
(Cercaria solemyae, Martin, 1944, and p. amphicteis,
Oglesby, 1961). In both cases, complete life cycles are
unknown and the organisms are questionably assigned to
the Sanguinicolidae. Until more complete life cycle infor-
mation is forthcoming, I think it is safest to assume
that the dorsal median fin fold is a derived structure

shared by the two families.

Character 10. Egg Type.
(0) Operculate, tanned shell.
(1) Non-Operculate, tanned shell.
(2) Non-Operculate, thin, elastic shell;
derived from state (0).
Operculate eggs with thick shells hardened by quinone
tanning (Cheng, 1973) are widespread among families in

the Strigeatida and among trematodes in general. The

 

non-ope]
appear '
substanl
shelled.
hatch t1
In cont
leave t
hatch 1

0f schis

non—Operculate eggs of schistosomes and sanguinicolids do not
appear to be homologous and their modes of hatching differ
substantially. Sanguinicolid eggs are elastic and thin
shelled. They apparently remain in the fish host and
hatch there, commonly in the gill filaments (Smith, 1972).
In contrast, thick, quinone—tanned eggs of schistosomes
leave the final host with the feces (and/or urine) and
batch in the external environment. I consider the eggs
of schistosomes and sanguinicolids as independently derived,
autamorphic characters because the details of miracidial
hatching suggest that the loss of an operculum may not
be homologous. In schistosomes, the top portion of the
egg comes off upon hatching in a nmnner analogous to the
opening of an operculum (Faust et al., 1934). In contrast,
the eggs of sanguinicolids appear to disintegrate upon
hatching, giving no suggestion of an operculumlike release

mechanism (Ejsmont, 1926).

Character 11. Tribocytic Organ.
(0) Absent.
(1) Present.

The tribocytic organ is a foliaceous or bulbous expan—
sion of the anterior end (Figure 8) unique to four (perhaps
more) families of strigeiform trematodes. Because of struc-
tural similarity, this complex organ is considered by
Dubois (1938) to be homologous in the groups possessing

it. I consider its presence a derived character.

 

 

 

 

 

 

Figure 8. Adult strigeiform trematode displaying the

tribocytic organ.

 

 

 

condit

Character 12. Condition of Intestinal Cecae.
(O) Terminate blindly.
(1) United at posterior end.

Intestinal cecae which terminate blindly (Figure
9A) constitute the widespread condition in families of
the order Strigeatida as well as in digenetic trematodes

.in general and are assumed to be the primitive condition.
The cecae of all schistosomes and one genus of the morpho-
logically diverse family Spirorchidae unite posteriorly
to fomn a continuous loop (Figure 9B). 1 have coded the
family Spirorchidae for the primitive condition, under
the rationale that while partially retaining the primitive
condition, one genus has independently developed the derived

morphology in parallel with the schistosomes.

Character 13. Anterior Cecal Branches.
(0) Absent.
(1) Present.
Anterior cecal branches (Figure 9C) are most likely
a derived condition unique to the Sanguinicolidae. Anterior
cecal branches appear to be absent in two poorly known
genera, which may eventually prove to be congeneric.
Chimaerohemecus trondheimensis has been reported twice
(Van der Land, 1967 , and Drenske, 1968) from the rat—fish,
Chimaera monstrosaq in the North Sea, while Orchispirium
heterovitellatum has been reported once (Madhavi and Rao,

1967) from the ray, Dasyatis imbricatus in the Bay of

 

 

 

 

 

 

 

Figure 9. Condition of intestinal cecae.

A. Terminate blindly
B. United at posterior end

C. Anterior cecal branches present

 

 

as previously thought. There is some indication that
anterior cecal branches may exist in reduced form in Orchi-
spirium, which has a bulbous expansion on either side
of the point where the esophagus meets the cecal bifur—
cation. A firm hypothesis of reduction and loss of the
branches in these two genera must await further information

and direct examination of specimens.

Character 14. Separation of Sexes.
(0) Monoecious.
(l) Dioecious.

Although dioecious species are rarely found in several
trematode families (e.g., Didymozoidae) among the Strigea—
tida, only the Schistosomatidae possess separate sexes.
In addition, this is the only trematode family in. which
all species are dioecious. It is likely that the dioecious
condition arose independently in a number of trematode

families and its OCcurrence in the schistosomes represents

a uniquely derived condition within the Strigeatida.

Character 15. Esophageal Gland Cells.
(0) Absent.

(1) Present.

 

spiriun

very I
studie:
cells

inap:

Charact

71

With the exception noted below, esophageal gland

families
Spirorchidae and Schistosomatidae and are coded as a derived

character. Gland cells have been reported in the two

aberrant sanguinicolids mentioned in the discission of

Character 13: Chimaerohemecus trondheimensis and Orchi-
spirium heterovitallatum. These two blood flukes are

very poorly known and until more detailed morphological

studies are available, I will assume that esophageal gland

cells of sanguinicolids represent one cu‘ two derivations

in a parallel with those of other blood flukes.

Character 16. Acetabulum (Ventral Sucker).
(0) Present in at least some life history
stage.
(1) Absent in all life history stages.

The acetabulum is found in nearly all digenetic trema—
tode taxa. In the Schistosomatidae the acetabulum may
be absent in adult worms (e.g., Gigantobilharzia) but
its presence in cercariae indicates loss during Inetamor-
phosis into the adult stage. The acetabulum is never

found in the sanguinicolids and I consider this a derived

loss character.

Character 17. Genital Pore.

(0) Male and female ducts open into a common

pore.

(1) Separate, monoecious.

 

 

 

 

 

Figure 10. Esophageal gland cells present.

 

 

 

A
ducts
Strigea
sanguin
are no
the se
both E
indivi<
somes

to th

the a
are 10
end 0;

indepe

74

(2) Separate, dioecious; derived from
state (0).

A common genital pore receiving both male and female
ducts is found in nearly all trematode taxa and in all
Strigeatida with the exception of the schistosomes and
sanguinicolids. The arrangements seen in these two families
are not likely to be homologous for two reasons. First,
the sexes are separate in the Schistosomatidae, while
both sets of reproductive organs are found in the same
individual in the Sanguinicolidae. Secondly, in schisto—
somes the genital pore of each sex is located anterior
to the reproductive organs, usually immediately behind
the acetabulum. In sanguinicolids, the genital pores
are located behind the reproductive organs, at the posterior
emd of the body. I consider these two conditions to be

independent derivations from the primitive state.

Phylogenetic Analysis

I conducted an analysis of the phylogenetic relation-
ships of families in the suborder Strigeata in order to
determine the out-group of the Schistosomatidae. Because
of the ambiguities present in data from the family Sanguin—
icolidae, I analyzed two different data sets in which
the problematic characters received different codings.
The derived state of Character 4 involves the loss of
the redial stage. Among families of the Strigeata, only

the clinostomes unambigiously possess rediae. Several

 

redial
species
Convers
colids
In
5 desi;
Once a
sketch:
5). '
as pre
In the
are p]

lies.

75

questionable reports suggest that a redial stage may be
present in certain sanguinicolid life cycles (see Character
4). Therefore in the first data set (Data Set I, Table
3), I scored the sanguinicolids as possessing the primitive
redial stage, under the assumption that the majority of
species have lost the redia in parallel with other families.
Conversely, in Data Set 2 (Table 3), I score the sanguini—
colids as having undergone a derived loss of the redia.

In a similar manner, the derived state of Character
5 designates loss of the pharynx in cercariae and adults.
Once again, a pharynx is reported as present in a few
sketchy reports of sanguinicolid cercariae (see Character
5). Therefore I have coded the sanguinicolid pharynx
as present in the first data set and absent in the second.
In the first data set, those sanguinicolids with no pharynx
are presumed to have lost it in parallel with other fami—
lies.

Both data sets listed in Table 3 consisted of 17
characters in a total of 37 states, designated for 8 fami-
lies. The components specified by the character distri-
butions are listed in Table 4. I have omitted single
taxon components specified by autapomorphies from Table
4 because they do not aid in producing branching diagrams.
To perform a component analysis, I used Felsenstein's
largest clique program to generate all possible combinations

of the components contained in Table 4. Computer analysis

 

 

76

TABLE 3

CHARACTER STATE DISTRIBUTIONS OF THE STRIGEATA

omcflpmeomowmfinom

odoHQOHOMHQm

odofiaoofioflswsdm

ocoflpdEOpmocflfio

odoflaapooonammo

ochunEOpm
IOHQHpOMOpOHQ

oonpdEOPonQHQ

oopHonnpm

DATA SET 1
Character
State

0*

0*

101
102
11
12
13
14
15
16

171
172

*Data Set 2 differs in substituting the derived state for

the starred characters.

 

r4

 

Component

 

77

TABLE 4

DISTRIBUTIONS OF THE STRIGEATA

Families

All families

Strigeidae, Diplostomatidae,
Proterodiplostomatidae,
Cyathocotylidae, Sanguinicolidae,
Spirorchidae, Schistosomatidae

Strigeidae, Diplostomatidae,
Proterodiplostomatidae,
Cyathocotylidae, Spirorchidae,
Schistosomatidae

Strigeidae, Diplostomatidae,
Proterodiplostomatidae,
Cyathocotylidae

Sanguinicolidae, Spirorchidae,
Schistosomatidae

Clinostomatidae, Sanguinicolidae,
Spirorchidae, Schistosomatidae

Spirorchidae, Schistosomatidae

Clinostomatidae, Sanguinicolidae

COMPONENTS SPECIFIED BY CHARACTER STATE

Defining
Character
State

1

4
(Data Set 2)

4
(Data Set 1)

(Data Set 2)
7
5, 8, 15
(Data Set 1)
8, 15
(Data Set 2)

9

—~———————_—__________—_

of Data
Set 2 y
Th

are ill
a quadr
Due to
have i
native
were c

featur

0f the

group,

78

of Data Set 1 generated two unique cladograms while Data
Set 2 yielded three topologies.

The cladograms generated by the component analysis
are illustrated in Figure 11. All five cladograms contain
a quadrichotomy consisting of the four strigeiform families.
Due to my character state designation, all four families
have identical codings. I chose to accept this uninfor-
mative coding scheme because the strigeiform families
were of interest only as potential outgroups. One other
feature common to all five cladograms is the designation
of the spirorchids as the sister taxon, and thus the out-
group, of the Schistosomatidae.

In Figure 11A and B, I assigned character state distri-
butions to the two branching diagrams produced by Data
Set 1; each contained 21 evolutionary steps. In Cladogram
1 (Figure llA), Component 6 unites the clinostomes and
the three blood fluke families on the basis of the shared
cercarial head-organ (Character 7). Component 8 defines
the clinostomes and sanguinicolids as sister groups because
their cercariae possess a dorsal-median fin fold (Character
9). The tetrapod blood flukes (Schistosomatidae and Spiror-
chidae) form a terminal dichotomy defined by Component
7. Two character states contribute to Component 7: 8--

derived penetration gland pattern and 15--presence of

esophageal glands.

In depicting the clinostomes and sanguinicolids as

a sister lineage to the schistosome-spirorchids, Cladogram

 

 

 

79

Figure 11. e Strigeata. Components
circled numbers. Gain of

a derived character state is indicated by

the symbol (X), reversal by (0). Optimized
branches are indicated by dashed lines.
A. Cladogram 1, Data Set 1, 21 steps

B. Cladogram 2, Data Set 1, 21 steps
C. Cladogram 3, Data Set 2, 22 steps
D. Cladogram 4, Data Set 2, 22 steps

E. Cladogram 5, Data Set 2, 22 steps

.onents
tin of
:ed by

timized

 

 

80

 

\:

 

 

 

 

 

81

 

 

 

 

 

icolid
l dif
place

family

of Da
that

to a
strig

the c

 

82

1 requires a parallel loss of the redial stage (Character
4) in ink; strigeiform families and tetrapod blood flukes.
This branching sequence could also be explained by a rever—
sal in which the redia reappeared in clinostome and sanguin—
icolid life cycle. The relationships portrayed in Cladogram
l differ markedly from standard interpretations which
place the three families of blood flukes in a single super-
family, the Schistosomatoidea.

The second cladogram produced by component analysis
of Data Set 1 also differs from traditional taxonomy in
that the clinostome—sanguinicolids form a sister group
to a branch containing the tetrapod blood flukes and
strigeiform families (Figure 118). Component 3 removes
the clinostomes and sanguinicolids since these taxa retain
rediae (Character 4) and cercarial pharynx (Character
5). Cladogram 2 requires the parallel evolution of the
cercarial head-organ (Character 7) in the clinostomes and
blood flukes, or alternatively its loss in the strigei-
form families.

The three cladograms generated from Data Set 2 each
require 22 evolutionary steps. Cladograms 3 (Figure 11C)
is identical in topology to Cladogram l, and the defining
components are the same in each case. Two conflicting
character state distributions are indicated. First, a

reversal to development in rediae (Character 4) occurs
in the clinostomes and second, the pharynx is either lost

in parallel in the sanguinicolids and tetrapod blood flukes

 

 

 

 

the c

in tr
redia
Sohic

L

ing

Clint

acte‘

equa

83

or a reversal in which the pharynx reappears (Character
5) occurs in clinostomes.

Cladogram 4 (Figure 11D) approaches a more traditional
view of blood fluke relationships. Once again, Component
6 groups together the clinostomes and blood flukes, but
this time the blood flukes branch as a monophyletic lineage
defined by Component 5. Here, derived loss of the cercarial
pharynx (Character 5) designates a component which corre-
sponds to the superfamily Schistosomatoidea. Two incon-
gruent character state distributions occur in Cladogram
4. A reversal to redial development (Character 4) is
indicated for the clinostomes and the dorsal-median fin
fold (Character 9) either evolves in parallel in the clino-
stomes and sanguinicolids or arises in Component 6 and
is subsequently lost in tetrapod blood flukes.

In Cladogram 5 (Figure 11E), Component 2 separates
the clinostomes as the sister taxon of the other families
in the suborder Strigeata based on a derived loss of the
redial stage (Character 4). Once again, the superfamily
Schistosomatoidea is designated by Component 5. The branch—
ing sequence of Cladogrmn 5 requires the parallel origin
Of two character states in both the sanguinicolids and
clinostomes: Character 7-—cercaria1 head-organ and Char—

acter 9--dorsal—median fin fold.

Both data sets produced alternative cladograms of
equal length, thus eliminating parsimony as a criterion

for choosing preferred cladograms. Instead I followed

 

 

a modif
Edwards
derived
case of
(Figure
are all
cercari
sanguiu
flukes
contra:
lies a:
losses
other

struct

iu -

Droh

84

a modified interpretation of the arguments of Hecht and

Edwards (1977), favoring cladograms defined by the fewest

derived loss characters. My preferred topology in the
case of each data set was the same: Cladogranl l and 3
(Figure 11A and C). In both cladograms, the clinostomes

are allied with the blood flukes on the basis of the derived
cercarial head—organ. Similarly, the clinostomes and
sanguinicolids group separately from the tetrapod blood
flukes as a result of the dorsal-median fin fold. In
contrast, cladograms which designate the blood fluke fami-
lies as a monophyletic lineage do so on the basis of derived
losses of 'the pharynx and redia. In addition, cladograms
other than 1. and 3 require parallel evolution of complex
structures such as the head-organ and dorsal—median fin
fold, further reducing the confidence in them.

My analysis therefore suggests that the blood flukes
are not a monophyletic lineage and thus brings current
taxonomy (If the superfamily into question. This analysis
is based on only 17 characters, though, and additional

information is necessary in order to settle the matter.

Phylogenetic Implications and
Host Relationships

Although the precise relationships of the families
in the suborder Strigeata are not entirely clear due to

problematic accounts in the literature, the results of

my aha

nomy a

famili
stome‘
edly
12),
first
the :
fluke
stoma
super
to (

Stud}

 

85

my analysis do pose questions with regard to current taxo-
nomy and phylogenetic history of the group.

In the preferred cladograms the four strigeiform
families cluster together as a sister group to the clino—
stome-blood fluke lineage. This arrangement differs mark-
edly from the prevailing concept of these taxa (Figure
12), reviewed most recently by Short (1983). LaRue (1957)
first proposed the subordinal ranking Strigeata to include
the superfamilies Schistosomatoidea (containing the blood
flukes), Strigeoidea (the strigeiform families), and Clino-
stomatoidea (the family Clinostomatidae); the latter two
superfamilies are generally considered more closely related
to one another than to the blood flukes (Figure 12).
Studying LaRue's account, I have been unable to discover
a specific morphological basis for his uniting the clino-
stomes and strigeiform families. I suspect that the classi-
cal arrangement of these taxa is based on life cycle pat-
terns and an unstated desire to keep the ”distinctive”
and ”advanced” blood flukes differentiated from other
trematodes. In my analysis, the clinostomes and blood
flukes are united (x1 the basis of 21 derived modification
of the anterior end of the cercariae known as the head-organ
(Character 7). Clinostome and blood fluke cercariae are
quite similar in most aspects of their morphology, although
the characters contributing to this similarity (Characters
2 and 3) are scored as primitive in my analysis, and there-

fore not indicative of derived relationship. If, as

 

 

 

 

 

 

 

 

 

 

 

Strig

Sirige

Figure 12. Classical interpretation of relationships
of the superfamilies of the Strigeata (after
Short, 1983).

 

Strigeoidea Clinostomatoidea

 

Strigeiforrn ‘
Families Clinostomatidae Sanguinicolidae Spirorchidae Schistosomatidae

onshiPS

(after

 

 

 

Brooks
stome:
and

in f:
be th

the c

ment
somes
thin]
toge
simi
Syna
are
of 1
cerr
acts

turi

88

Brooks et a1. (1983) suggest, the blood flukes (and clino-
stomes) are a paedomorphic lineage, then Characters 2
and 3 may well be secondary simplifications which are
in fact synapomorphic in nature. If this turns out to
be the case, the argument for a close relationship between
the clinostomes and blood flukes would be strengthened.

One consistent trend in all cladograms was the place-
ment of the spirorchids as the sister group to the schisto-
somes. Although this arrangement does reflect classical
thinking, prior systematists have grouped the two taxa
together on the basis of host relationship and life cycle
similarity, rather than on shared morphological features.
Synapomorphies which unite the schistosomes and spirorchids
are penetration gland pattern (Character 5), presence
of esophageal gland cells (Character 15), absence of the
cercarial pharynx (Character 8) and loss of rediae (Char—
acter 7). This result establishes the blood flukes of
turtles as the out-group which I will use in my analysis
of the phylogenetic relationships of schistosome genera.

Having chosen the out—group of the Schistosomatidae,
I examined hypotheses suggested by the preferred cladogram
concerning historical patterns of host relationships.
In Figure 13, host taxa are assigned to the terminal
branches of the preferred cladogram. My discussion ‘will
focus upon the host relationship patterns of the
clinostome-blood fluke alliance. As a result of an

extensive phylogenetic analysis of the phylum

 

 

 

 

 

 

 

 

89

Figure 13. Vertebrate host distributions assigned to

the preferred cladogram depicting relationships

of the Strigeata.

 

 

 

90

 

uoEGEomofiEom 89:80st

m_oEEoE .923 33:58 $52 3533

82025398 89.6E2856

5.: 68692.25

22.6820

5: .3253 .mss agate

[Ed 130
onships

mopwmtm Suzceofioafiocmaoi oou..EE2mo_a5 8238286

3.88. .fi: .EEEE

moxocm £55388
.mEEEoE .225 "2.22:8

 

 

 

Platyh
comple
Digene
deriva
For 6
the V
outsr
found
It a
Strig
and (
primf
then
one
the
to
the
at

flu]

91

Platyhelminthes, Brooks et al. (1983) suggested that
complex life cycles are 21 defining characteristic of the
Digenea and that simpler patterns represent internal
derivations resulting from losses within specific lineages.
For example, the metacercarial stage which encysts upon
the vertebrate intermediate host is distributed widely
outside of the Strigeata, but within the suborder it is
found only in the clinostomes and strigeiform families.
It appears most parsimonious to suggest loss within the
Strigeata rather than multiple derivations in other families
and orders. If in fact the three-host life cycle is found
primitively in the clinostomes and strigeiform families,
then the preferred cladogram (Figure 13) indicates that
one vertebrate host has been lost in the life cycles of
the three blood fluke families. Unless a reversal back
to a three-host life cycle occurred, in the clinostomes,
the preferred cladogram suggests that there have been
at least two separate losses of vertebrate hosts in blood
fluke life cycles: one in the sanguinicolids and at least
one in the tetrapod blood flukes.

Host distribution patterns portrayed in Figure 13
indicate that the ancestors of the Strigeata were inhabiting
either modern tetrapods or their primitive ancestors prior
to the splitting off of the spirorchid-schistosome branch,
which now penetrate the definitive host directly. If
one loss event accounts for the two-host life cycle in

both spirorchids and schistosomes, then the loss is indeed

 

 

 

 

aucie
offsh

rise

final
have
host

on i

foun
in ‘
thre
noti
Sch
the

and

 

 

92

ancient because turtles are thought to be a very early
offshoot from the primitive reptilian stock which gave
rise to other tetrapods.

Conversely, the sanguinicolids may have lost the
final or definitive host. Ancestral sanguinicolids may
have altered to penetrate their intermediate vertebrate
host and develop directly rather than encysting in or
on it.

Some support for this latter hypothesis is perhaps
found 111 differences :hi the mechanics of host involvement
in the three blood fluke families. The cercariae of all
three families penetrate the host directly, but Virtually
nothing is known about this process other than in the
Schistosomatidae. The way in which the eggs return to
the external environment differs between the sanguinicolids
and tetrapod blood flukes. In both the schistosomes and
spirorchids, the eggs leave the vertebrate host in the
feces, urine, or very rarely in a nasal discharge. In
the sanguinicolids which have been studied, the eggs travel
to the gill filaments where the miracidia hatch and pene—
trate to the exterior (Smith, 1972). These differences
may reflect separate historical patterns of host inhabi-
tation, or alternatively, they may reflect a long history
Of divergence accompanied by specific modifications for
differing host environments. Specifically, the struc-
ture of the piscean circulatory system may require exit

routes which differ from those needed in tetrapods.

 

 

 

 

hos

med
co]
are
re]

ad<

br:
at
f 1‘

me

d1

1‘!

 

93

An alternative and perhaps less parsimonious hypothesis
could view the current piscean hosts of the sanguinicolids
as the ancestral definitive host. In this case, the primi-
tive life cycle would have involved two fish as vertebrate
hosts, and the intermediate fish would have been eliminated.

Although I favor the first hypothesis which suggests
that the sanguinicolids lost the definitive vertebrate
host while the tetrapod blood flukes eliminated the inter-
mediate vertebrate host, further information on sanguini-
colid biology is needed before more definite conclusions
are made. A detailed analysis of phylogenetic and host
relationships of the strigeiform families might provide
additional data pertinent to this discussion.

The phylogenetic implications of intermediate inverte—
brate host relationships of the Strigeata are tentative
at best since the life cycles of non-schistosome blood
flukes are so poorly known. Figure 14 depicts the inter-
mediate host distributions represented by the preferred
cladogram of familial relationships. Larval development
has been reported for at least some members of all families
as occurring in the strictly freshwater molluscan class
Pulmonata. Sanguinicolid-like cercariae have been reported
from non-gastropod hosts in marine environments, partic-
ularly annelids and pelecypods (Smith, 1972). Unfortu-
nately, complete life cycles are unavailable and therefore
definite identifications are lacking. Spirorchids are

recorded only from pulmonate snails, but since adults

 

 

 

 

 

94

Figure 14. Intermediate invertebrate host taxa assigned
to the preferred cladogram depicting relation-

ships of the Strigeata.

 

,ssigned

.1ation-

.4

Pulmonata

Pulmonata

Diplostomatidoe Proterodiplosintmfidae

Cyathocotylidae

         
    

Strigeidae

Clinostomatidae

95
Pulmonata Pmm
Poiecypoda (?) I Pulmonata Prosobranchia ‘1
Amelido (?) * Prosobranchia Opisthobrmctio l‘
Sanguinicolidae Spirorchidae Schistosomatidae

Pulmonata

Pulmonata

 

 

Pulmonata
Prosobranchia

  

{ occurs in this taxon in less than
|O°/o of reports

 

 

of many species are found in nmrine turtles, prosobranch
gastropods and perhaps non-gastropod invertebrates must
also be involved.

In a manner similar to that used in character analysis,
I utilized Farris optimization to assign intermediate
host taxa to interior nodes of the preferred cladogram.
Despite the limitations imposed by incomplete data, certain
generalizations can be drawn from intermediate host distri-
butions. Figure 14 indicates that pulmonate gastropods
are the primitive hosts of the Strigeata. Since pulmonates
are considered derived relative to prosobranchs (Purchon,
1977), this implies that up to three independent invasions
of prosobranchs may have occurred in the history of the
Strigeata. The occurrence of the primitive intermediate
host taxon in the life cycles of all three families of
blood flukes suggests that at least one independent inva-
sion of prosobranch gastropods has occurred in each family.
This may be related to the invasion of the marine habitat
by these groups, because although prosobranchs are found
in fresh water, they are presumed to have originated in

salt water, where the majority of taxa now reside (Purchon,

1977).

Summary

A definite hypothesis of the relationships within
the Strigeata is not available at present, but my results

favor cladograms which group the clinostomes and

 

 

 

 

 

 

97

sanguinicolids and differ significantly from classical
thought in breaking in) the superfamily Schistosomatoidea.
Cercariae of the sanguinicolids and clinostomes possess
a complex dorsal—median fin fold which unites the two
families. Most the cercariae positively assigned to both
families are virtually identical, with the questionable
exception of the absent pharynx in some sanguinicolids.

If the pharynx and redial stage do turn out to ‘be
present in some sanguinicolids, this would further support
separating the sanguinicolids from the tetrapod blood
flukes. Brooks et a1. (1983) use the loss of the redial
stage in) unite the tdood flukes and strigeiform families.
A cladogram defined by this derived loss character appeared
in my analysis, but I preferred branching diagrams supported
by the presence, rather than absence of morphological
characters.

Alternatively, if the pharynx and redia are absent
in sanguinicolids, then this absence would support the
traditional view of the blood flukes as a monophyletic
group. In any case, the blood flukes are presently united
solely on the basis of derived loss characters, and any
such monophyletic grouping must postulate the parallel
evolution of the synapomorphic dorsal-median fin fold
in the clinostomes and sanguinicolids.

My analysis indicates that three—host life cycles
involving pulmonate molluscs and vertebrate intermediate

and definitive hosts are primitive characteristics of

 

98

the Strigeata. Although the monophyletic nature of the
Schistosomatoidea is open to question, my analysis did
show that the tetrapod blood flukes are sister groups,
with the Spirorchidae forming the out-group of the
Schistosomatidae. Inconsistent species descriptions and
incomplete life cycle data constrain further speculation
at the present time. Detailed cercarial descriptions
will most likely be the key to clarifying ambiguous
character codings and thus the family relationships of

the Strigeata.

1..

 

 

In

[C3

W_M_m.-l ,_ ,1

THE SCHISTOSOMATIDAE

Survey of Characters

The 24 characters used in my analysis of the Schisto—
somatidae are presented below, followed by discussions
of schistosome phylogeny and host relationships. The
explanation of terminology presented at the beginning
of the survey of Family Level Characters (page 47) also

applies here.
Genus Level Characters

Character 1. Testes Shape.
(0) Compact.
(1) Lobulate.

Within the out-group, the Spirorchidae, the testes
occur in a variety of forms ranging from globular to elon-
gate and spiraled (as the familial name implies). In
all cases, the testes occur as discrete units with a more
or less solid outline. In a similar manner, schistosomes
generally possess discrete, roughly ovoid testes (Figure
15A). An exception to this pattern occurs in the genus

Schistosomatium where the testes are lobulated and diffuse

(Figure 15B). Thus in contrast with the discrete testes
displayed by spirorchids and most schistosomes, the
99

 

 

 

 

 

Figure 15. Shape of testes.

A. Compact

B. Lobulate

 

 

 

 

 

 

 

 

 

 

102

lobulated testes of Schistosomatium appear to represent

a derived, autapomorphic state.

Character 2. Testes Location.

The blood flukes of turtles and the majority of dige-

netic trematodes possess testes which lie between the

cecal branches (Figure 16A). In all schistosomes, the

cecal branches unite posteriorly to a. greater or lesser

degree to form a common cecum. The testes are located

I will therefore consider the condition in which testes

lie alongside the common cecum to be derived relative

to testes which lie between the cecal arms.

Character 3. Dendritic Branches on Paired Cecae.
(0) Absent.
(1) Present.

Among blood flukes, dendritic branches on the paired
cecae (Figure 17) are found only in the genera Schisto—
somatium and Heterobilharzia. I assume that the condition
is homologous, although more detailed developmental and
histological data would be necessary to confirm this.
In both genera, the condition is dispayed more prominently

in the female. A similar condition occurs, presumably

 

 

 

 

 

‘ Figure 16. Location of testes.
A. Between cecal branches

B. Along common cecum

 

 

 

 

 

 

 

Figure 17.

Dendritic branches present on paired cecae.

 

 

 

 

(A

r

n

106

 

 

 

 

 

 

 

107

in parallel, in the clinostome genus Clinostomoides (Yama-

guti, 1971).

Character 4. Gynacophoric Canal.

(0) Absent.

(1) Present, running the length of the body.

(2) Present, surrounding the genital pore
only; derived from state (1).
(3) Present, restricted to the posterior

one—half of the body; derived from state
(1).
The gynacophoric canal is formed in male schistosomes

by an infolding of the sides of the body. This character-

istic structure is found only in schistosomes and gives

the family its name. It is absent in the flat-bodied
genus Dendritobilharzia which .I will conservatively score

for the primitive condition since there is no obvious

loss. The remaining schistosome genera can be placed

in one of three categories, depending upon the extent

of development of the canal. In the most simple arrange-
ment, e.g., Schistosoma, the body walls fold inward for

much of the length of the worm (Figure 18A). This may
be accompanied by muscular thickening to a greater or
lesser degree; this muscularization is apparently used
to keep pairs intact while in the blood vessels of hosts.

Variation in the extent of muscularization suggests further

 

 

 

108

Figure 18. Form of Gynacophoric Canal.
A. Running the full length of the body
B. Surrounding genital pore only
C. Restricted to the posterior one-half of

the body

 

 

e—half of

 

 

 

 

 

llO

refinement of this character, but histological data are
needed for confirmation. I consider this state of the
canal to be primitive relative to the following two canal
arrangements. I feel that this is a relatively safe assump-
tion, since the following genera all possess derived body
forms not represented elsewhere in the Schistosomatidae
or Spirorchidae.

The first of these two derived canal arrangements
occurs in the genera Trichobilharzia and Gigantobilharzia.
In these two highly elongate genera, the canal is repre-
sented fur a short anterior muscular expansion surrounding
the genital opening (Figure 18B). In the genus Schisto-
somatium, the canal is limited to the posterior one—half
of the body (Figure 180) in correlation with the derived
position of the male genital pore (see Character 10).
In the absence of further information as to the development
of the canal in these three forms, I consider the latter
two to be independent derivations from the simple canal
which runs the length of the body.

The genus Bilharziella, is somewhat problematic since
reports on the length or actual presence of the canal
vary. Khalifa (1972) reports that in ,B. polonica, the
edges of the body have a tendenCy to roll inward and that
the actual canal is short and located in the region of
the testes. Lal (1937) reports a well developed canal
running the length of the body in B. indica. Based on

these reports and my own observations that suggest some

 

 

 

 

 

111

folding of the body wall, I think it is safest to score
this genus for the widespread, simple gynacophoric canal.
Interpretation of the length of the canal can vary according
to degree of relaxation and state of preservation, and
since there are reports indicating the presence of a full
canal (Lal, 1937), I will assume it is present until clear

evidence to the contrary is presented.

Character 5. Adult Oral Sucker.
(0) Present.
(1) Absent.

The oral sucker is present in all members of the
out-group family, Spirorchidae, and in xnost schistosomes.
I therefore consider its absence a derived loss. The
oral sucker is always absent in members of the genus Dendri-
tobilharzia. It is usually reported as absent in the
genus Gigantobilharzia, but reports vary. Najim (1956)
reviews the problem and observes that the oral sucker
is present in one or both sexes in at least g. monocotylea
and ,g. gyrauli. Brackett (1942) reported it absent in
g. lawayi, but it is clearly present on one specimen
(#44862) deposited by Brackett in the U.S. National Helminth
Collection. In light of these conflicting reports, it
is evident that the primitive state occurs more than once

in the range of variation shown in the genus Gigantobil-

harzia. I have therefore scored the genus as primitive

 

 

 

 

 

 

112

and assumed that the loss of the oral sucker has occurred

one or more times within the genus.

Character 6. Adult Acetabulum (Ventral Sucker).
(0) Present.
(1) Absent.

The ventral sucker, or acetabulum, is present in
cecariae in all schistosomes and spirorchids for which
the life cycle is known. It is absent in the adults of
two genera: Gigantobilharzia and Dendritobilharzia. It
is apparently lost during the process of penetration and
maturation. I consider the absence of the acetabulum

to be a derived loss character.

Character 7. Ovary Shape.
(O) Ovary globular, or weakly coiled at
the anterior end.
(1) Ovary weakly coiled throughout.
(2) Moderately coiled ovary; derived from
state (1).
(3) Highly coiled ovary; derived from state
(2).
(4) Coiled ovary with prominent lateral
loops; derived from state (2).
In its most widespread form in the out-group and
in the order Strigeatida as a whole, the ovary is a simple
globular structure with a relatively smooth outline. I

have therefore labeled this simple, globular ovary shape

 

 

 

 

 

113

Figure 19. Ovary shape.
A. Globular or weakly coiled

Weakly coiled throughout

B
C Moderately coiled

D. Elongate and highly coiled
E

Coiled, with prominent lateral loops

 

 

 

 

 

 

 

115

as the primitive character state. Within the Schisto-
somatidae, the ovary is either simple and globular or
more or less coiled to form a spiraled structure. Support
for the suggestion that the simple ovary represents the
primitive state comes from developmental studies. The
detailed ontogenetic study of Schistosomatium douthitti
by El-Gindy (1951) shows that the ovary forms first as
a globular structure and then elongates the coils during
the final stages of development. In the genus Schistosoma,
ovary shape is variable ranging from pyriform in S. japoni-
gum_ to elongate with a slight ”torting" of the anterior
end in ,S. mansoni (Price, 1929). Basch (1981) reports
that when sexual maturation of S, mansoni is incomplete,
a pyriform ovary similar to the one seen in S. japonicum
is retained. These studies imply that spiraling of the
primitive globular ovary occurs as a terminal addition
to the development sequence. Coiling occurs to variable
degrees in all genera of schistosomes and below I will
explain my coding of character states based on degree
of coiling.

In the genera Schistosoma and Orientobilharzia, certain
species have simple globular ovaries while all other genera
show some degree of coiling (see Dutt and Srivastava,
1961, for a review of ovary morphology in the Schisto—
somatidae). Although problematic, the most conservative

coding for the two genera is for the primitive state with

independent derivations of a partially coiled ovary in

 

 

 

 

 

116

certain species of both genera (Figure 19A). The genus
Bilharziella. is highly consistent in displaying an ovary
which is only slightly coiled with one major coil running
the whole length of the ovary (Figure 19B) (Kowalewski,
1886; Khalifa, 1972, etc.). I have coded this condition
for derived state (1): Ovary slightly coiled.

The majority of schistosome genera possess ovaries
that are moderately coiled (Figure 190) consisting of
four cn' more coils packed tightly together (e.g., Austro-
bilharzia). Although there is variation in the length
and degree of coiling, I have scored schistosomes with
moderately coiled ovaries as possessing the same character
state. It is likely that the degree of coiling in the
report of a given species would be affected by the age
of the worm as well as its ability to mature in a particular
host species. Lack of information on individual variation
makes further refinement of this character unreasonable
at the present time. Further research on growth and
development in a range of species might yield useful infor—
mation.

Despite the above mentioned ambiguity, two extreme
variations indicate that separate codings are in order.
In two genera (Trichobilharzia and Ornithobilharzia),
the ovary is extremely elongate and loosely coiled (Figure
19D). I scored this condition as an independent derivation
from state (2) because sparse developmental data indicates

elongation during ontogeny (McMullen and Beaver, 1945).

 

 

  
  
  
  
  
  

117

In the genera Heterobilharzia and Dendritobilharzia, the
coiled ovary is thrown into a series of prominent lateral
loops (Figure 19E). I coded this condition as independently
derived from state (2) until additional developmental

information is available that might suggest otherwise.

Character 8. Arrangement of Vitelline Follicles.
(O) Follicles scattered between and/or outside
of paired cecae.
(l) Follicles paired along both paired cecae.
(2) Follicles paired along common cecum;
derived from state (0).

In the spirorchids and other families of the Strigeata,
the vitelline or yolk producing glands are scattered in
a diffuse manner between or lateral to the two cecal
branches (Figure 20A). This arrangement occurs in a number
of schistosome genera and I considered it to be primitive.
Two additional arrangements also occur and I scored these
as independently derived from the primitive condition.
The genus Bivitellobilharzia derives its names from the
paired arrangement of the Vitellaria (Figure 20B) which
surround each of the cecal branches (Vogel and Minning,
1946). This distinctive arrangement is also found in
the genera Schistosomatium and Heterobilharzia. Although

the paired arrangement is not as distinct as in the
published report of Bivitellobilharzia mentioned above,

it is clearly evident in developing Schistosomatium as

 

 

 

 

 

Figur

 

 

 

Figure 20.

Arrangement of vitelline follicles

laria).

A. Scattered

B. Paired along both paired cecae
C. Paired along common cecum

D. Unpaired

(=vitel-

 

 

119

 

 

 

120

portrayed by El-Gindy (1951) and in the specimens of
Heterobilharzia that I examined (U.S.N.H.C. #39420).

Another equally distinct arrangement is seen in the

genera Schistosoma, Ornithobilharzia, Austrobilharzia,
and Orientobilharzia. In these genera, the vitellaria

form prominent lobes which are paired on either side of
the common cecum (Figure 20C). Although the length of
the common cecum may vary, the vitelline arrangement is
quite consistent.

A fourth vitelline pattern consisting of single,
large vitellaria arranged in tandem (Figure 20D) has been

reported at least three times in the literature. One

of these reports, that of Faust (1924) concerning Ornitho-

bilharzia odhneri, has generated considerable controversy.

Dutt and Srivastava (1961) erected the new genus Sino-

 

bilharzia to house this species based on the peculiar
vitelline structure and a few inconsistencies in genital
arrangement. Farley (1971) considered 9, odhneri to be
an aberrant member of the genus Austrobilharzia. Although
the reported genital arrangements have their own problems
(discussed later), subsequent authors have failed to point
out the inconsistencies between Faust's illustration and
his written description of the vitellaria. Faust's Figure
2 portrays a female 9. odhneri with large ovoid vitelline
follicles spanning the width of the body and is the basis

for Dutt and Srivastava's (1961) taxanomic action. In

contrast, on page 551, Faust reports ”the paired vitelline

 

 

 

 

 

 

121

follicles occupy a large portion of the body from the
posterior position to the sub-distal region" (emphasis
added). Faust stated that he based his description on
a total of nine male and four female worms. There were
no females present in the type material I received from
the U.S. National Helminth Collection (#60990), which
consisted 13f nearly all. the reported specimens (fragments
made the exact number difficult to determine).

I suspect that the written description is correct
and that an error was made in illustrating the material.
Given that the number of male specimens which I examined
nearly matches the total number of worms on. which Faust
based his description, I speculate that the testes of
a male worm were mistakenly illustrated as vitellaria.
From my own observations, the testes are large and span
the width (if the body, resembling closely the illustrated
structures. This species has not since been reported
in the literature.

Two other reports describe schistosomes with a single
row of vitellaria. In his description of Ornithobilharzia
filamenta, McLeod (1940) stated that ”the vitellaria, are
in the form of a single row of numerous relatively large
follicles, each of which is almost as wide as the body.”
The photomicrographs which accompany the report are not
particularly revealing, and no mention is made of the
relationship of the vitellaria to the common cecum. Since

no further note is made of this aberrant arrangement,

 

 

 

 

122

I can only conclude that the paired vitellaria have been
interpreted as a single structure, a description which
seems plausible if the common cecum did not stain promi-
nently enough to demarcate the two lobes.

The final report of an avian schistosome with an
aberrant vitelline structure comes from an incomplete
note published by Young (1937), involving an unidentified

schistosome from the marbled godwit (Limosa fedoa). The

author describes the vitellaria as, ”occupying about 2/5
of the body, . . . somewhat pyriform in shape and filling
most of the diameter of the worm." The description. was

based (n1 a single pair of worms, and these specimens were
destroyed during sectioning before a positive identification
could be made. Although these specimens appear to belong
to the genus Ornithobilharzia, no special attention was
given to the peculiar arrangement and once again no refer-
ence to the common cecum was made.

As in the previous two accounts, the information
in Young's report is sufficiently incomplete that an error
in interpretation of such limited material cannot be ruled
out. Until more substantial information is available,
I am not willing to recognize a separate character for
these questionable reports of aberrant vitelline structure.
For the purpose of analysis, I have included EL odhneri
under the separate genus Sinobilharzia, a placement which

differs from Farley (1971) who included the species in

the genus Austrobilharzia. This listing is based on Faust's

 

 

 

123

report of genital differences, not on vitelline structure.
These genital differences have their own associated problems

and I will discuss these in a later section (Character

14, below).

Character 9: Dendritic Branches on Common Cecum.
(0) Absent.
(1) Present.
Dendritic branches from the common cecum (Figure
21) appear to be autapomorphous relative to the out-group
and are restricted to the genus Dendritobilharzia. The
life cycle of this genus is unknown and thus information

pertaining to the development of these branches is unavail-

able.

Character 10. Position of Testes and Genital Pore.

(O) Testes and genital pore immediately
post-acetabular.

(l) Testes and genital pore located in
the xniddle of the body, not associated
with the acetabulum.

(2) Testes located in the posterior end
of the body; genital pore immediately
post-acetabular; derived from state
(0).

In schistosomes, the testes are located either between
the cecal branches or along the length of the common cecum

(see Character 2). In the former situation, the testes

 

 

 

 

 

 

Figure 21. Presence of dendritic branches on the common

cecum of Dendritobilharzia.

 

 

 

 

 

 

126

and genital pore are located immediately behind the aceta-
bulum in the majority of genera (Figure 22A). Two excep-
tions to this pattern occur. In the genus Schistosomatium,
the genital pore and testes are located in the middle
of the body, well behind the acetabulum (Figure 223).
In the genus Heterobilharzia, the testes are found in
the posterior end of the body and are connected by a long
vas deferens to the genital pore which is situated imme-
diately behind the acetabulum (Figure 220).

Developmental data presented by El-Gindy (1951) show
that in Schistosomatium the testes and terminal geni-
talia form early in development between the cecal branches
immediately behind the acetabulum. During subsequent
development, the whole complex moves posteriorly and comes
to lie in the middle of the body. Thus the early develop-
mental stages of Schistosomatium resemble the final adult
stage of the majority of genera, in which the testes lie
between the cecae. Based on this information, I suggest
that the pattern seen in Schistosomatium is derived relative
to other schistosomes.

Although developmental data are lacking for Hetero-
bilharzia, it seems likely that a third developmental
pattern is involved and I have scored it as independently

derived from the primitive, intercecal condition.

 

 

 

 

 

 

 

 

 

Figure 22.

127

Position of testes and genital pore.

A.

Immediately post—acetabular

Located in the middle of the body

Testes located in

the body;

acetabular

genital

the

pore

posterior end of

immediately post

 

 

 

128

 

t
S
0
p
y
l
e
t
a
.1

 

 

 

129

Character 11: Position of Female Genital Pore.

(0) Immediately behind the cecal bifurcation
and acetabulum.

(1) Genital pore opens at extreme anterior
end, anterior to cecal bifurcation
and acetabulum, when present.

In the spirorchids and nearly all schistosomes, the
genital pore lies posterior to the cecal bifurcation and
acetabulum, where present (see Character 10, Figure 22A).
In tWO genera, Dendritobilharaia and giganighilharaia,
the female genital pore lies anterior to the cecal bifurca-
tion and close to the oral opening (Figure 23). Based

on out-group comparison, I assumed this to be 21 derived

state.

Character 12. Number of Eggs in the Uterus.
(0) One to ten eggs.

(1) More than fifty eggs.

Trematodes in general are noted for their high repro-
ductive capacity which is often evidenced by enormous

egg output by individual females. Among the strigeiform

trematodes, there is a tendency toward the production

of a few ((10) relatively large eggs. The spirorchids

always have one or, at the very most, a few ((5) large

eggs present in the uterus. The widespread occurrence

Of a single egg in the uterus of most blood flukes will

therefore be considered the primitive state. Members

 

 

 

 

 

 

130

Figure 23. Position of female genital pore in Dendrito-

bilharzia and Gigantobilharzia.

 

 

 

 

 

u 26111319;

 

 

 

 

 

 

 

132

of at least five schistosome genera have sac-like uteri
packed with many (SO—100+) eggs. This is true for all
species of the four genera (Macrobilharzia, Heterobilharzia,
Schistosomatium and Dendritobilharzia) I have scored as
possessing the derived state. Species in the genus Schisto-
soma generally contain a single egg, but some members
(e.g., .§. japonicum) may contain over a hundred. Since
the primitive state is present in the range of variation
seen in Schistosoma, I assume that the presence of multiple
eggs in some species represents one or more independent
derivations of the state. That multiple egg production
in the genus Schistosoma is not homologous with that of
other schistosome genera is also indicated by the arrange-
ment of the eggs in the uterus. In S, japonicum, for
example, the eggs are loosely packed, generally in tandem,
in a very long, straight uterus. In the other genera,
the eggs are densely packed in an enlarged, sac-like uterus

which tends to fold and bend along its length.

Characters 13—17 report on details of genital morpho—
logy, which are generally the most difficult morphological
traits to observe using conventional straining techniques.
Information on many of these characters is not present
in a substantial number of descriptions. The majority
Of authors fail to explain whether or not they feel a
structure is truly absent or whether their technique 'was

insufficient to distinguish presence or absence. In my

 

 

133

own experience, certain structures (e.g., prostate cells)
are extremely difficult to distinguish on many mounted
worms; it is clear that histological examination would
have been much more useful. In most cases, histological
data were not available and I had to rely on inconsistent
literature descriptions and mounted worms stained by
researchers other than myself.

Faced with this inconsistent data, I needed to formu—
late a standard method of coding problematic genital char—
acters. I decided to use a synthesis of all available
data on a given genus. I took information from all avail-
able species descriptions, as well as from museum specimens.
In recording the presence or absence of a structure, I
assumed that it was more likely to be overlooked in a
description than to be reported as present when it actually
was not. Therefore, if a given structure was reported
in. at least two species descriptions (H' was observed by
myself in museum material, I coded it as present. My deci-
sions were influenced by the quality of a given description
and/or museum specimen, since certain authors provided
much more detail in their accounts than others. Overall,
I am satisfied with this approach and feel it gives a
reasonable picture of the morphology of each genus; at
worst, this approach provides hypotheses to be examined
by future workers. Problems arose Inainly when a generic

designation relied on a single description (e.g., Sino-

 

bilharzia and Bivitellobilharzia). These and other

 

 

 

 

 

 

134

significant problems are discussed under individual charac-
ters.

I used a similar argument in determining transformation
series in genital Characters 13-17. All of the characters
are present in some form in the spirorchids and most other
families of the Strigeata. These structures show a
bewildering pattern of variation in presence or absence
in digenetic trematodes as 21 whole, as well as in their
sister group, the Monogenea. Although numerous and highly
specific genital autapomorphies exist, it seems unlikely
that the presence of common genital structures (e.g.,
prostate, seminal receptacle, cirrus, etc.) which serve
the same function and display similar morphologies repre-
sents independent derivation in so many lineages. I have
therefore scored absence of these genital characters as
derived from the schistosomes, in the hope that a biologi-
cally meaningful pattern of derived loss states will be

revealed.

Character 13. Prostate Gland Cells.
(0) Present.
(1) Absent.

Well developed prostate gland cells surround the
intromittent organ (Figure 24) in all members of the sub-
families Bilharziellinae and Gigantobilharziinae (Farley,
1971). They are present in reduced form in the genus

Dendritobilharzia and were reported only in the most recent,

 

 

 

 

 

 

 

 

 

 

 

 

Figure 24. Cirrus surrounded by prostate cells.

 

  

. . II..I.—

 

 

 

137

highly detailed analysis of this genus, presented by Ulmer
and VandeVusse (1970). In another relatively recent report,
prostate cells were found by Short and Holliman (1961)
in Austrobilharzia penneri, although they are not mentioned
in prior works on this genus. As mentioned earlier, I
chose to separately analyze 01d and New World forms assigned
to the genus Macrobilharzia. Prostate cells were recorded
from African (Fain, 1953), and Indian (Baugh, 1963) forms,
but not from the poor accounts of North American (Price,
1931) and South American (Kohn, 1964) worms. There is
one report of prostate cells in Schistosoma japonicum
(Tada, 1928) but because both the species and genus are
so well studied and prostate cells are never again reported,
I consider this an error. Prostate cells appear to be
absent in all other genera. They are never mentioned
in any works on the genus Heterobilharzia and since I
was not able to observe them on mounted slides (U.S.N.H.C.

#14572 and #39420), I assume that they are absent.

Character 14. Cirrus.
(0) Present.
(1) Absent.

The intromittent organ, or cirrus (Figure 24), is
definitely absent in only two genera: Schistosoma and
Orientobilharzia (Price, 1931, and Dutt and Srivastava,
1961). In these well-studied genera, the seminal vesicle

terminates at the genital pore and an eversable cirrus

 

 

 

 

 

 

 

138

is absent. I have tentatively scored the cirrus as absent
in Bivitellobilharzia because it is not nwntioned 111 the
extremely vague report of Vogel and Minning (1940). In
general, the male genital arrangement of Bivitellobilharzia
resembles that of Schistosoma, but the genus has been
reported only twice and until further data are available,
I think the given coding is most appropriate.

Although the intromittent organ is not mentioned
by Faust (1924), I scored the cirrus as present in Sipg;
bilharzia. I did so because the cirrus pouch described
by Faust was clearly visible in museum specimens (U.S.N.H.C.

#60990), and in all other schistosomes which possess a

cirrus pouch, the cirrus is also present.

Character 15: Male Genital Complex.

(0) Cirrus pouch, internal and external

seminal vesicle present.

(1) Cirrus pouch and coiled internal seminal
vesicle present.

(2) Cirrus pouch very small, surrounding
terminus only; external seminal vesicle
present; derived from state (0).

(3) Cirrus pouch and a relatively small,

straight seminal vesicle present; derived

from state (0).

 

 

 

139

(4) Cirrus pouch absent; seminal vesicle
small and globular; derived from state
(0).

The male genital apparatus is a complex set of organs,
and the interrelationship of the constituent parts is
unclear. In addition to the prostate and cirrus, the
following structures may be found :Ui a number of combi—
nations: cirrus pouch, internal seminal vesicle (within
the cirrus sac), external seminal vesicle (outside the
cirrus sac), and seminal vesicle proper (in the absence
of the cirrus sac). The evolutionary relationships of
these component parts are unclear; for example, in the
absence of histological data, it is currently impossible
to determine if the seminal vesicle proper represents
an internal or external seminal vesicle or combination
of both after the loss of the cirrus pouch. Instead of
recording the presence and absence of individual structures,
I recorded the male terminal genitalia as a complex since
at least five distinct patterns occur. Since all the
above mentioned structures are found in the out-group,
I coded as primitive those genera in which all are present.
States 1-4 consist of different combinations of these
structures and are coded as independent derivations. It
is likely that a more informative pattern of relationship
exists but until considerably more information is available,

the approach I have taken seems most conserative.

 

 

 

 

Mp—ii

139

(4) Cirrus pouch absent; seminal vesicle
small and globular; derived from state
(0).

The male genital apparatus is a complex set of organs,
and the interrelationship of the constituent parts is
unclear. In addition to the prostate and cirrus, the
following structures may be found in a number of combi-
nations: cirrus pouch, internal seminal vesicle (within
the cirrus sac), external seminal vesicle (outside the
cirrus sac), and seminal vesicle proper (in the absence
of the cirrus sac). The evolutionary relationships of
these component parts are unclear; for example, in the
absence of histological data, it is currently impossible
to determine if the seminal vesicle proper represents
an internal or external seminal vesicle or combination
of both after the loss of the cirrus pouch. Instead of
recording the presence and absence of individual structures,
I recorded the male terminal genitalia as a complex since
at least five distinct patterns occur. Since all the
above mentioned structures are found in the out-group,
I coded as primitive those genera in which all are present.
States 1-4 consist of different combinations of these
structures and are coded as independent derivations. It
is likely that a more informative pattern of relationship
exists but until considerably more information is available,

the approach I have taken seems most conserative.

 

 

 

 

 

 

 

140

Character 16: Laurer's Canal.
(0) Present.
(1) Absent.
Laurer's canal is a structure of undetermined function
which is found in female trematodes. It was formerly

thought to be a rudimentary vagina, but evidence supporting

this hypothetis is lacking. Laurer's canal is absent
in Schistosoma and Orientobilharzia. It is not mentioned

in any description of Macrobilharzia and although I feel
the data are not conclusive for this genus, I coded its
absence as a derived loss. The canal is reported as present
in at least some representatives of all other genera,
and is consistently reported as present in all genera

in subfamilies other than the Schistosomatinae.

Character 17: Seminal Receptacle.
(0) Present.
(1) Absent.

Although it is seldom mentioned, the seminal receptacle
(Figure 25) has been reported for the genus Schistosoma
on more than one occasion (i.e., Faust et al., 1934; Sinha
and Srivastava, 1956). The only group in which the seminal
receptacle is consistently reported as absent is the genus
Orientobilharzia. Although previously mounted specimens
which I examined showed no trace of the structure, a vaguely
suggestive object was observed in one specimen of p. turke—

stanicum which I stained and mounted (U.S.N.H.C. #45820).

 

 

 

 

 

141

Figure 25. Size and shape of seminal receptacle.
A. Simple and sac-like

B. Enlarged; long and sinuous

 

 

 

 

 

 

 

143

Although I have coded its absence as a derived loss in
Orientobilharzia, further study on more extensive material
may reveal the presence of the seminal receptacle in the

genus.

Character 18. Size and Shape of Seminal Receptacle.
(0) Simple and sac-like.
(1) Englarged; long and sinuous.

When present in spirorchids, the seminal receptacle
is always small and globular. It is described in two dif—
ferent ways, as either an expansion of the germiduct,
or as a separate sac—like outpocketing from the duct (Price,
1934; Mehra, 1939). In schistosomes, the seminal receptacle
is usually a small, globular sac arising from the germiduct
(Figure 25A). I have scored the enlarged and convoluted
seminal receptacle which occurs in three genera (Tricho-
bilharzia, Gigantobilharzia and Dendritobilharzia) as

derived condition (Figure 25B).

Character 19: Spatulate Projections of Posterior Extremity.
(0) Absent.
(1) Present.

Nearly all reports for the genus Trichobilharzia
depict a modification of the posterior end of the body
(Figure 26) which is variously described as Spatulate,
fan-like, and bulb or club-like. I suspect the extent
to which the projections are manifested is dependent upon

the degree of contraction of an individual worm during

 

 

 

 

 

 

Figure 26. Spatulate projections of posterior extremity

of Trichobilharzia.

 

extremity

 

 

 

 

lot

bot
des
in

Ne:
b1]
me]

an

ar

th

Ct

146

preservation. It is generally reported as present in
males, but McMullen and Beaver (1945) report it in both
sexes for T, physellae, I, ocellata and l} stagnicolae.
It is possible that further examination of this large
and complex genus will yield useful information pertaining
to interspecific variation in the size and shape of the
posterior extremity.

In his description of the subfamily Gigantobilharziinae
(genus Gigantobilharzia only), Yamaguti (1971) mentioned
lobe—like lateral projections of the posterior end in
both sexes. This is apparently due to Odhner's original
description of the genus Gigantobilharzia (Odhner, 1910)
in which a modification of the posterior end was reported.
Nearly all subsequent reports of the genus describe a
blunt or tapering terminus. Lal (1937) and Farley (1963)
mention dilated or swollen posterior ends on .Q- egreta
and .Q- lawayi respectively, but at least Farley clearly
distinguished the condition he observed from the Spatulate
structures seen in Trichobilharzia. Likewise, my own
limited observations indicate that Spatulate modifications

are restricted to the genus Trichobilharzia. I have scored

their occurrence as an autopomorphy.

Character 20. Location of Male Genital Pore.
(O) Seminal vesicle relatively simple;
genital pore located just posterior

to the acetabulum.

 

 

 

ac<

ti<
so]
am
tht
any
Ve

Sll

 

147

(l) Seminal vesicle elongated; genital
pore well behind the acetabulum but
anterior to the cecal reunion.

(2) Seminal vesicle elongated; genital
pore well behind the acetabulum and
cecal reunion; derived from state (1).

Although the seminal vesicle is somewhat enlarged
in a. few spororchids, it never achieves the enlarged and
elongate condition seen in some members of the Bilharziel-
linae and Gigantobilharziinae. In the genera Trichobil-
harzia and Gigantobilharzia, the seminal vesicle is
extremely long and the genital pore lies well behind the
acetabulum and/or cecal reunion (Figure 27B). This situa-
tion differs from the general condition displayed by
schistosomes; sparse developmental information in McMullen
and Beaver (1945) indicates that early development of
the genital primordia occurs just behind the acetabulum
and that the genital pore moves posteriorly as the seminal
vesicle grows and lengthens. This developmental sequence
suggests that the condition just described is derived
relative to the arrangements seen in the Schistosomatinae,
where the genital pore lies just posterior to the acetabulum
(Figure 27A). A somewhat intermediate condition coded
as such, is displayed by the genera Bilharziella and
Dendritobilharzia. In these genera, the seminal vesicle
is moderately enlarged and the genital pore lies between

the cecal branches, anterior to the cecal reunion. This

 

 

 

 

 

 

 

Figure 27.

 

Location of male genital pore.

A.

B.

Immediately post-acetabular

Well behind acetabulum and cecal reunion

 

 

 

 

Ch

 

 

 

150

coding consists of a somewhat bold hypothesis, and more
detailed developmental data would provide a stronger basis

for the hypothesis of relationship.

Character 21. Length of Common Cecum; Male.
(0) Common cecum short; fused only at poster-
ior end.
(1) Cecae fused for at least half of the

body length.

Character 22: Length of Common Cecum; Female.
(0) Common cecum short, fused only at poster—
ior end.
(1) Common cecum long and straight.
(2) Common cecum long and sinuous; derived
from state (0).

In the out-group, cecal branches usually remain unfused
at the posterior extremity. Stunkard (1923) reported
a few instances of fusion in the normally unfused genus
Spirorchis. Sporadic occurrences of fused cecae may be
related to the elongate body form characteristic of blood-
inhabiting flukes. In schistosomes,the cecal arms always
fuse but the relative length of the fused cecum varies
in two ways. First, the paired cecae may fuse only at
the posterior extremity, resulting in a very short common
cecum (Figure 28A). I consider this condition to be primi-
tive, since fusion occurs late in development and results

in an arrangement similar to that seen in the above

 

 

 

 

 

 

 

 

Figure 28.

 

Length of common cecum.

A. Short
B. Long and straight

C. Long and sinuous

 

 

 
 

 

 

 

 

 

10.
th
th

in

de

06

DE

Ck

153

mentioned aberrant spirorchids. The second pattern involves
a long common cecum which runs at least half the length
of the body (Figure 28B). In this case, the cecal branches
fuse early in development and subsequent elongation occurs
along the common cecum. Cecal length may vary between
the sexes of a given genus; therefore I have used separate
characters for male and female worms.

In addition to variation in length, females with
long cecae can be further subdivided into a group in which
the cecun1 is straight (Figure 28B) and a. group 1J1 which
the cecum is sinuous (Figure 28C). Limited ontogenetic
information (McMullen and Beaver, 1945; Cort, 1921; and
Faust et al. 1934) suggests that fusion occurs later in
development inthe sinuous group than in taxa with straight
cecae. I have tentatively coded these two states as inde-

pendently derived, until I obtain more informative data.

Character 23. Condition of the Acetabulum.
(0) Simple.
(l) Areolated.

In spirorchids and the majority of schistosomes,
the acetabulum is a simple, sucker-like structure. Excep-

tions to this occur in reports of both African and Asian

Macrobilharzia. The acetabulum of the Indian form is
described by Baugh (1963, p. 306) as follows: "the ventral
sucker has a scalloped margin -- this is a consequence

of its peripheral area being marked into a series of

 

 

 

 

 

 

te
ei
01

0]

CI

areolae, usually seventeen or eighteen in number. This
areolated periphery of the ventral sucker is an outstanding
morphological and distinguishing character of the fluke.”
Fain (1955, p. 357) described the acetabulum of the African
form as possessing ”17 a 22 festons." The presence or
absence of this autapomorphic character in American forms
is unclear; since areolae are not mentioned, I have scored
the acetabulum as simple. In an intriguing but inconclusive
report, Kohn (1964) pictures this genus with segmented
acetabulae. Since no special reference is made in the
text, this ”segmentation” may just be pictured for artistic
effect, emphasizing the three dimensional nature of the
organ. Once again, further information is necessary in

order to clear up this ambiguity.

Character 24: Testes Number.
(0) 2.
(1) >2, <10.
(2) >10, <30.
(3) >30, (100.
(4) >100, (200.
(5) (200.

The presence of two testes is the pattern most fre-
quently encountered among digenetic trematodes. Spirorchids
usually have two testes, but numbers may vary from one
to many. Schistosomes always have more than two, and

often up to many hundred testes. In scoring this character,

 

 

 

 

155

I assumed the condition of multiple testes was achieved
independently in the two above mentioned families, since
spirorchids possess the primitive condition of two ‘testes
within the range of variation displayed by the family.

In determining a 'transformation series based on the
number of testes within genera of the Schistosomatidae,
I took a number of factors into consideration. The number
of testes in :1 given individual varies according to the
age (Chu and Cutress, 1954) and size of the worm; the
latter is particularly true of the filiform genera. In
the introduction, I discussed the present knowledge of
factors (suitability of host environment, competition,
etc.) which influence growth and development, and thus
indirectly influence testes number. Unfortunately, it
is nearly impossible to determine the age of a sample
taken from a wild-caught host. Since most species descrip-
tions are based on samples taken from one or a very few
host individuals, the chance of encountering immature
worms may bias the estimate of testes number downward
for a given genus. In addition, it is impossible to assess
the suitability of a particular host species with regard
to parasite development when descriptions are based on
small samples.

Taking these factors into account, I based my estimate
of testes number on the upper two-thirds of the range
of variability shown by all species of a given genus.

This approach resulted in the six non-overlapping categories

 

 

 

 

IO

IO

156

listed at the beginning of this section, each derived
in sequence from the state immediately previous to it.
Genera with fewer testes ((100) showed less variability
than those with numerous testes, as might be expected.
I realize that in my attempt to compartmentalize data
on testes number, I run the risk of obscuring true natural
variability. Although subject to reevaluation in the
future, I feel my hypothesis of relationship is reasonable

at present.

Phylogenetic Analysis

The data matrix presented in Table 5 consists of
24 characters in a total of 64 states, designated for
14 genera._ For the purpose of computer analysis, I included
a hypothetical ancestor scored for the primitive state
of all characters. With two exceptions, I followed the
generic taxonomy of Farley (1971). The first exception
is in the genus Macrobilharzia, where I found enough incon-
sistencies in reports of Old and New World forms to merit
a separate analysis of each. Macrobilharzia is the most
poorly described avian taxon within the Schistosomatidae;
my separation of forms may simply represent the general
lack of knowledge of the group.
The second deviation from Farley's taxonomy is my
use of the name Sinobilharzia for his Austrobilharzia
odhneri, which had originally been described in the genus

Ornithobilharzia. As I discussed under individual

 

 

 

 

 

 

 

157

81%qu o o o o o o
—'[',EqOIlI-Ipu9(1 .-+ O o H r—4 H H r-t
etzreq o o o o o
_I.‘[q01u,3319 H H H O O H H H O
'etzanq OHOHHOOOr—l ooo
-ttqoqou.t H 7‘
m
a“ BIIGIZJ’BUIIH OHOHOOOOHOOOOO
H
E-'
<2
g sizarqttqouts OOOo—IOOOOI—lr—IOOOu—l
to
E3
tz ’3
f9, B'Jq OOOHOOOOHHHOOl—l
m —IIQOQ1IU-IO
8
m (pI‘IOMpIO) ooor-(ooooaaoooo
m s: BIZJ'QQIIQOJO‘BW
m E-«
e
53% (pI‘IOMMN) OOOHOOOOHI—IOOOO
<3 ’etzx'eqttq010'ew
E—« m
‘3 121211311 0 o o o o o
5': —IIClO-Ilsnv r-1 0 H .—( o o o .—+
D
E} umtiew o o o
E —osoistqos .—i r-I H H O O H r-l O r-i O
m
a "812.1%; 0
o Or—IOOOOOOOOOI—l
_ I o UGIJ
Li! 1.q 3 . O
E4
E, etzreq
m —IIq0-1913H ooafiooooaao—«HO
a:
:3 31211311
2: —I'.EQOII911A18 OOOr—(OOOOHq—IOOr—IO
n:
E
U 81110801811108 OOOr-IOOOOOOOOOr—l

H

0.)

PG)
04-)
(6:6
54+)
NV)
.53
0

HQ)” HNO’JQ‘r—(N
HNMV‘Q‘V‘LOQOBFFFww

 

 

 

H P4 H Pd H F4 H rd H La H rd H
m -Q o: m cu m cn A to M +4 o c: w
A co m: H N +4

OOHHOOOD—‘l—‘OOOOO

ooHoooHoHHOOOO
<3 0 re 0 c>+a o :3 o P4 o c><3 o
o :3 o c: o +4 o c>ra 0 <3 0 <3 O
o c> o C) o +4 o c: F4 o c: o c: o
o C) o c: 0 <3 0 <3 0 <3 0 c: o c>
h‘ 0 <3 0 <3 0 <3 0 <3 0 c: o c: o

t“ O C><3 o c><3 o c: o FJ<D o c

HOOOOOOOOHl—‘OOH

891

6191s
Jeioeinqo

Schistosoma

Bivitellobil—
harzia

Heterobil-
harzia

Orientobil-
harzia

Schistoso-
matium

Austrobil-
harzia

Macrobilharzia
(New World)

Macrobilharzia
(Old World)

Ornithobil—
harzia

Sinobilharzia

Bilharziella

Trichobil—
harzia

Gigantobil-
harzia

Dendritobil-
harzia

(penutiuoo) g etqel

 

 

Table 5 (continued)

arzan
—IIq0111PU90

BIZJEQ
—IIq01UBBID

BIZJBU
—ttqou0111

BtIeIZIBUIIa

EIZJBHIIQOUIS
BIZJBQ
—IIqu11UJO

(DIJOM PTO)
EIZJEQIIQOJOBW

(DTJOM MGN)
sizaeqttqoaoew

sizaeq
_Itq011snv

mutinw
—osoistqos

Btzxeq
—IIq01UGIJO

BIZJBU
_IIq0J918H

stzanq
—I1q01I911A18

emosoistqos

Character
State

19
201

159

202
21

221

222
23

 

 

HNMVID

24
24
24
24
24

pr(
he]
66
wh
21.
ha:
St
(N
fo
st
on
o
a

2

o

e

 

 

 

The components specified by character distributions
in the data matrix are listed in Table 6. I have not
included single taxon components specified by autapomorphies
in Table 6 since they do not contribute information useful
in generating branching diagrams. In order to perform
a component analysis, I used Felsenstein's largest clique
program to generate all possible combinations of the compo—
nents contained in Table 6. The computer analysis generated
66 unique cladograms.

I limited detailed analysis to the fourteen cladograms
Which contained Component 12, consisting of the taxa Bilhar—
ziella, Trichobilharzia, Gigantobiharzia and Dendritobil—
harzia. Component 12 is specified by four derived character
States, more replicates than occur for any other component.
(No Icomponents /are specified by three character states,
four are specified by two, and the rest by one character
state.) These derived states are: 21--testes located
On either side of common cecum; 201-—male genital pore
located well behing the acetabulum, 21-—common cecum in

male greater than one half of the total body length; and
222——female common cecum long and sinuous. In addition

to the number of replications, all four states represent

derived gains as opposed to losses, lending further

 

161

TABLE 6

COMPONENTS SPECIFIED BY CHARACTER STATE
DISTRIBUTIONS OF THE SCHISTOSOMATIDAE

\

Component

Genera

Defining
Character
State

________________________________________________________________

1
2

All taxa

Bivitellobilharzia, Hetero-
bilharzia, Orientobilharzia,
Schistosomatium, Austrobilharzia,
Macrobilharzia (N.W.), Macro-
bilharzia (O.W.), Ornithobil-
harzia. §inghilharzia, Bilhar-

ziella, Trichobilharzia, Gigan—
tobilharzia, Dendritobilharzia

Schistosoma, Bivitellobilharzia,
Heterobilharzia, Orientobil—
harzia, Schistosomatium, Austro-
bilharzia, Macrobilharzia (N.W.),
Macrobilharzia (O.W.), Ornitho-
bilharzia, Sinobilharzia, Bilhar-
ziella, Trichobilharzia, Giganto-
bilharzia

Schistosoma, Heterobilharzia,
Schistosomatium, Austrobilharzia,
Macrobilharzia (N.W.), Macro—
bilharzia (O.W-), 9221129211-
haraia, §inghilharaia..hilhar-

ziella, Trichobilharzia, Giganto-
bilharzia, Dendritobilharzia

Bivitellobilharzia, Hetero-
bilharzia, Orientobilharzia,
Macrobilharzia (N.W.), Macrobil-
harzia (O.W.), Ornithobilharzia,
Sinobilharzia, Bilharziella,

Trichobilharzia, Gigantobilharzia,

Dendritobilharzia

241

242

24

 

 

 

 

 

 

 

Table 6 (continued)

Component

6

10

11

12

13

14

Bivitellobilharzia,
bilharzia,
Austrobilharzia,

Genera

Hetero-
Schistosomatium,
Macrobilharzia

(N.W.), Macrobilharzia (O.W.),

Ornithobilharzia,
Trichobilharzia, Gigantobilharzia,

Sinobilharzia,

Dendritobilharzia

Schistosoma, Bivitellobilhar-
zia, Heterobilharzia,

 

Orientobilharzia,
Macrobilharzia (N.W.)

w.

Ornithobilharzia,
Heterobilharzia,

Schisto-

Sinobilharzia,
Schistosomatium,

Macrobilharzia (N.W.), Macro-
bilharzia (O.W.), Dendritobil-

harzia

Macrobilharzia (N.W.), Macro-
bilharzia (O.W.), Trichobil-

harzia, Gigantobilharzia,
Dendritobilharzia

Schistosoma,
Austrobilharzia,
gn_obi_lh.2£zi_a

harzia,

Schistosoma,
Austrobilharzia,

harzia

Bilharziella,

Orientobilharzia,
graifhghil-

Orientobilharzia,
Ornithobil-

Trichobilharzia,

Gigantobilharzia, Dendritobil-

hm

Schistosoma,

Orientobilharzia,

Macrobilharzia (N.W.), Macro-
bilharzia (O.W.)

Macrobilharzia (N.W.), Macro—
bilharzia (O.W.), Gigantobil-
harzia

Defining
Character
State

72

13

12

24

21, 22

16

24

 

 

 

 

 

Table 6 (continued)

Component

15

16

17

18

19

20

21

22

23

24

Genera

Trichobilharzia, Gigantobil—
harzia, Dendritobilharzia

Macrobilharzia (N.W.), Orni-
thobilharzia, Sinobilharzia

Schistosoma, Bivitellobil—
harzia, Orientobilharzia

Bivitellobilharzia, Hetero-
bilharzia, Schistosomatium

Heterobilharzia, Schisto-
w

Austrobilharzia, Macrobil—
harzia (N.W.)

Heterobilharzia, Dendrito-
bilharzia

Ornithobilharzia, Trichobil-
harzia

Gigantobilharzia, Dendrito-
bilharzia

Trichobilharzia, Dendrito-
bilharzia

Defining
Character
State

18

152

14, 15

 

 

 

Vai

am
Pai

gel

an
Th

ne

 

164

confidence to my decision to focus attention on cladograms
containing Component 12.

Three of the character states defining Component
l2 involve the cecum. I cannot rule out the possibility
that these states are in some way developmentally or func-
tionally correlated since detailed ontogenetic sequences
are unavailable for nearly all genera. I think it is
not unreasonable to treat each character state separately.
Variation 111 cecal fusion and testes location both within
and between genera suggest some support for my approach.
Patterns of cecal fusion differ significantly between
genera, and within a given genus, the cecum may be fused
in females and separate in males (e.g., Schistosoma),
fused. in both sexes (e.g., Bilharziella), or unfused in
both sexes (e.g., Macrobilharzia). Likewise, testes distri—
bution and number vary consistently, and variability is
particularly marked in genera in which the cecum is unfused.
Testes may be tandem and numerous (e.g., Orientobilharzia),
few and clustered midbody (e.g., Schistosomatium), numerous
and clustered terminally (e.g., Heterobiharzia), etc.
Thus a broad sample of comparative development data is

needed before the question of correlation among these
character states is resolved.

After narrowing my analysis to 14 cladograms based
on the replication of Component 12, I then assigned char-

acter state distributions to each branching diagram using

 

 

 

Iva

Id

“-6

rn

165

Farris optimization. The fourteen branching diagrams
are presented in Figure 29A-N. In these cladograms, groups
defined by actual components are illustrated 'using solid
lines, while optimized branches are indicated by dashed
lines. The total number of evolutionary changes in char—
acter states or steps is indicated along with each clado-
gram.

The fourteen cladograms can be viewed as two sets
of seven branching diagrams each, the sets differing only
in the placement of two taxa. In Cladograms 1-7, Component
24 defines a terminal dichotomy consisting of the genera

Trichobilharzia and Gigantobilharzia, while in Cladograms

8-14, Component 23 describes a terminal dichotomy of
Dendritobilharzia and Gigantobilharzia. In each case,

two character state changes are involved; thus exchanging
Trichobilharzia and Dendritobilharzia does not alter the
total number of steps in each corresponding pair of clado-
grams.

The character state distributions in the fourteen
resulting cladograms offered no clear avenues along which
to further narrow the field of choices. In the absence
of other criteria, I invoked parsimony to choose the clado-
gram which made the fewest statements about character
state change during evolution (Farris, 1983).

Two minimum length cladograms, each with an equal
number of steps, result from the use of parsimony: pre—

ferred Cladograms 2 and 9 (Figure 29B, I), with 64

 

 

 

 

 

 

Figure 29.

166

Component analysis of the Schistosomatidae.
Components are indicated by circled numbers.
Gain of a derived character state is indicated
by the symbol (X), reversal by (0). Optimized
branches are indicated by dashed lines.

A. Cladogram l, 71 steps

B Cladogram 2, 64 steps

C. Cladogram 3, 68 steps

D Cladogram 4, 66 steps

E. Cladogram 5, 67 steps
Cladogram 6, 69 steps

Cladogram 7, 77 steps

20 "1:1

Cladograms 8-14 correspond with Cladograms

1-7 but contain Component 23 in place

of Component 24.

 

 

_vN
m_
a 4 _¢
N... W
m
. 0
.o _.
Q.
m_
M ON N _m _NN
1 _ e ._
N w. s
\ a 10.0—
x
. m N_ m.
eszm: 9 N MN m c
w \ veN 9
MN VN
n
a 9
6m em _NN m. _NNee o.
_ m E. On. N
:0 N. N N. v.
neN wok _vo N m N_ .u i
6/0 /V( 0 ’0 9w «0
Ofivo 0,6. 00% «06¢. ’90 90.0 /O/OOA/ 4000‘! Anna} 6va
’9 ’0. or. Q 9 o 0 av 490v av 61. o«
a? 0%. b. I0. «0 #9 01A,? ’7 x0. ’9 00/ ’0,
.9 e or. o e 4 core a... 9. a. 9.
0/. 0/. 6v 0. Are f0 rev fo/evo $0 «00, eve
O. O O. O/ 0/ OAT 0/1991.
.1 b c m w...
M m M .1... o
m n n p .3
0 .1 O a
s n 1
+0.. N S. S C

 

 

168

 

 

 

 

 

169

 

meN
NvN
_vN
_s
_v
NN
__N
.8
N
N
e...
__
nNmVNo ma oNv
N_
NN
9 _ m Q
s m.
.9 vm M.
m n
N N: v. w. mVN
man 4.9 so 6v. 9o
492.. seen (too 90% 9%
9x. 9x. ’90. 94. x?
a o9. to Cop.
Onwv Oz. 0/1.

 

 

 

170

 

 

 

 

0/
43/
0
a,
0/
'6‘
0I
-<‘
_/
7’4) 9x " °°
0
«>90 m¢ X\
6‘ ”v
0’ — N
""0 "9’9...
69 (”90. 1 ‘ b“
04% ”a 9 r “* —
. ’0 u- N
0/ 994% — 9 N
450 0’0/ 9
a a
99/9]
.41, _
.9 a:
945°
9
4990
94, ,i
. O _
0% r0 _" '
O /"O 0;“ N O
.9 4, .19 o
“o .0 0
(”960/ N
0/ o E |D
.g, o ‘
0
9’9 4’ — _Nn"
0% Na.) 9 N
o 0t»
//9
O 'lelo N
"‘50 “”9
‘7/
900% _ V
09 V \0 lg
0/ 0 . 0’
.eJ
O
9
[/9090 (9'05
/ = 9
0, 'V ' '
euloy Lav
{/9 N

23

\
\
\
\
\
sf
9
car '0“.
N N
2 9
“3:

171

I3

 

 

 

 

 

//
'9
avg, N _
”a N
'90449 n
0% 91> . 3
-' O
h
96‘; Nsrm ‘9on _9
4'06, N g, N N
’0 rs
a, _
.40 _Ni
0 “do
{/41 _oo O‘x\£'g
o ,0 _
Q/o-Io [x ('11 e
N I")
:2 2 ~ —
9/ n‘ “X~ “_7
"I900, ‘2, N 9
p «5‘ 3 cu
'VO/OJ N ‘2
0,? 40°
409 4' N '2
(’90 v N 713 —
49/9 [x 9 IO
’9 _
o
, _ o
9%? g
N ‘o
|\ N
1'" ' e ‘”
009 'srln rs mm to: e
0 ° — 9 9‘)
9.090 ,0
(’04, WHEN No
~( ' v N

 

 

173

 

 

 

 

 

'h

 

 

 

 

 

 

175

evolutionary steps in each. I will discuss the implications
of the character state changes described by these branching
diagrams and then briefly compare them to the other 12
cladograms for which state changes were analyzed.

The preferred cladograms each contain a basal tricho-
tomy consisting of one branch composed of genera assigned
to the subfamilies Bilharziellinae and Gigantobilharziinae,
a second clade including Old and New World members of
the genus Macrobilharzia, and a final branch which involves
all other genera of the Schistosomatinae.

The relationships indicated by the basal portion
of the preferred cladograms differ somewhat fronl conven-
tional interpretations of schistosome subfamilies. Genera
of the Bilharziellinae and Gigantobilharziinae are generally
considered more closely related to one another than to
members of the Schistosomatinae; therefore placement of
their constituent genera on the same branch is not sur—
prising. Within the branch defined by Component 12, how-
ever, the arrangement of genera brings into question the
traditional division into two subfamilies. Also, the
genus Macrobilharzia branches separately from the rest
of the genera assigned to the subfamily Schistosomatinae,

where it has been traditionally placed.

In Cladogram 2, Gigantobilharzia, Trichobilharzia
and Dendritobilharzia are separated from the primitive
genus Bilharziella, and fornx a component defined by three

character states: 181--seminal receptacle is long and

 

 

176

sinuous, 244--greater than 100 but less than 200 testes,
and an optimized character: ll--female genital pore opening
at the extreme anterior end. Component 24 defines a ter—
minal dichotomy consisting of the genera Trichobilharzia
and Gigantobilharzia. Two derived character states describe
Component 24: 42—-gynacophoric canal limited to the region
of the genital pore and 202--genital pore located behind
a greatly enlarged seminal vesicle. Character 11 reverses
to the primitive state in Trichobilharzia, resulting in
the genital pore opening behind the acetabulum. Because
of the freedom Farris optiminzation allows in interpreting
the nature of character state changes, the derived state
of Character 11 can alternatively be viewed as a parallel
derivation in Dendritobilharzia and Gigantobilharzia.
Either intepretation results in the same number of steps.
The terminal dichotomy that appears in Cladogram
2 differs from the traditional interpretation of the sub-
family Gigantobilharziinae (Farley, 1971). Trichobilharzia
rather than Dendritobilharzia shares a most recent common
ancestor with Gigantobilharzia. In contrast, Cladogram
9 incorporates Component 23 which is composed of the two

genera Farley (1971) places in the Gigantobilharziinae:

Gigantobilharzia and Dendritobilharzia. Two character
states defined Component 23: 6--derived loss of the adult

acetabulum, and ll—-female genital pore located at the

extreme anterior end.

 

 

 

 

 

 

177

The nonparsimonious character state distributions
indicated by Component 23 can be interpreted in two ways.
In one case, Dendritobilharzia is defined by the loss
of two derived character states found in, Trichobilharzia
and Gigantobilharzia: Complete loss of the gynacophoric
canal (Character State 42) and reversal to a shortened
seminal vesicle (Character State 202). Under the alter-
native interpretation, both Character States 42 and 202
arise in parallel in Trichobilharzia and Gigantobilharzia.
In either case, the total number of evolutionary steps
remains the same and until further data are available,
I can find no basis to choose a single preferred branching
diagram between Cladograms 2 and 9.

In preferred Cladograms 2 and 9, the genus Macro-
bilharzia branches independently of the Schistosomatinae.
Optimization resolves a portion of the multichotomy present
after component analysis by uniting Old and New World
Macrobilharzia iJ1 a single clade. Although Old and New
World forms appear similar, much of their resemblance
occurs in primitive character states. Interestingly,
four optimized states define a component that unites the

genus. Two of these states (244, 245)

refer to the
extremely large number of testes that Macrobilharzia shares
in parallel with most members of Component 12. One
character refers to the large number of eggs in the uterus

(Character 12) and is shared in parallel with the genera

Heterobilharzia, Schistosomatium and Gigantobilharzia.

 

 

 

I?

178

The final optimized character state (l6--derived loss
of the Laurer's canal) is also found in parallel in
Schistosoma and Orientobilharzia. I consider the absence
of the Laurer's canal in Macrobilharzia questionable since
it is overlooked in many reports and the few accounts
available for the genus are sketchy at best. Additional
descriptions will hopefully clarify the status of
Macrobilharzia relative to other genera.

Remaining genera traditionally assigned to the Schisto-
somatinae are united in an optimized branch combining
Components 10 and 18 on the basis of a derived loss of
the prostate complex. This is a problematic scoring since
the prostate is often hard to detect and a reversal in
which the prostate appears apparently occurs in the genus
Austrobilharzia. The prostate is also lost in New World
Macrobilharzia or was perhaps overlooked in the few existing
descriptions.

To date, there have been no systematic studies examin—
ing the relationships of genera within the Schistosomatinae.
In discussions of the subfamily, mammalian and avian genera
are usually treated separately; therefore it is interesting
to note that my analysis separates mammalian genera in

two components, one of which contains avian genera as

well. Schistosomatium, Heterobilharzia and Bivitello-
bilharzia comprise Component 18. These three mammalian

taxa share a derived vitelline structure found no where

else among the Strigeata. It consists of a set of vitelline

 

 

 

 

 

179

glands on either side of both paired cecae (Character
State 81). The Afro-American genus Bivitellobilharzia
is one of the most poorly known schistosomes and details
of genital morphology, which in my analysis arise in paral—
lel with Schistosoma and Orientobilharzia, are vaguely
described in the literature.

Component 19 appears in all of the 14 initially pre-
ferred cladograms and consists of the monotypic genera
Heterobilharzia and Schistosomatium. These two North
American genera are united on the basis of three characters.
Both share a state in which the paired cecae possess lateral
diverticulae (Character 3) and both have a nearly identical
arrangement of the male genitalia (Character State 153)
which consists of a simple cirrus pouch and a straight
seminal vesicle. One optimized character (Character 12)
describes Component 19 (ml the basis of numerous eggs in
the uterus.

A different vitelline arrangement, paired glands

on either side of the common cecum (Character State 82)

describes Component 10. This component consists of the
genera Sinobilharzia, Ornithobilharzia, Austrobilharzia,
Orientobilharzia and Schistosoma. Optimized Character

State 152 describing male genital morphology also defines
Component 10. Component 11, consisting of Ornithobilharzia,
Austrobilharzia, Orientobilharzia and Schistosoma nests
within Component 10, and is defined by Character State

221-—female common cecum long and straight. The

 

 

 

 

 

 

 

180

quadrichotomy produced by Component 11 is resolved during
optimization. Loss of Character State 152 creates an
optimized branch containing the trichotomy Austrobilharzia,
Schistosoma and Orientobilharzia. Finally, an optimized
component containing Schistosoma and Orientobilharzia
is described by three derived loss character states (14,

154, 16) and two reversals (7l

, 72) which result in the
primitive state «of Character ’7 occurring in 'the terminal
dichotomy.

In addition to Cladograms 2 and 9, I assigned character
state distributions to the 12 other cladograms which con-
tained Component 12 (Figure 29A, C—H, J-N). The number
of evolutionary steps contained in each of these cladograms
ranged from 66 to 77. Differences in the patterns of
the 12 additional cladograms reflect the inclusion of
components incompatible with those present in Cladograms
2 and 9.

Eight (Cladograms 1-4, 8—11; Figure 29A-D, H-K) of
the total 14 cladograms are similar to one another in
possessing Component 18, made up of the nmmmalian genera
Schistosomatium, Heterobilharzia and Bivitellobilharzia.

When Component 18 is absent, four cladograms (Cladograms

5 and 12, 6 and 13) group Bivitellobilharzia with Schisto-

soma and Orientobilharzia. These three genera make up
Component 1?, defined by two character states: 154--male

terminal genitalia simplified by derived loss, and 14——

derived loss of the cirrus. Thus Component 1? groups

 

 

 

 

181

together the predominantly Old World mammalian schistosomes,
keeping separate the North American genera found in mammals.

Cladograms 2 and 9 differ from the other 12 cladograms
in that they include Component lO, composed of Sinobil-
harzia, Ornithobilharzia, Austrobilharzia, Schistosoma
and Orientobilharzia. The absence of Component 10 allows
for the inclusion of a. number of other components which
significantly alter cladistic relationships. In Cladograms
l and 8, an inclusive component (defined by Character
States 71--slightly coiled ovary, and 72-—moderately coiled
ovary) separates the genera Schistosoma and Orientobilharzia
as basal branches. This placement contrasts markedly
with Cladograms 2 and 9, where terminal placement of the
two genera requires the reversal to more primitive states
of a number of characters, including Character 7. In
Cladograms l and 8 (Figure 29A and H) as in all cladograms
which separate Old and New World Macrobilharzia (Cladograms
l and 8, 3 and 10, 5 and 12, 6 and 13, 7 and 14), the
number of parallel steps is increased. A major basal
separation in (Hadograms 3 and 10 is forced by Component
7, which separates Austrobilharzia and Old World Macro-
bilharzia from other genera of the Schistosomatinae on
the basis of Character l3--derived loss of the prostate.
Component 13 clusters the Old and New World Macrobilharzia
with the genera Schistosoma and Orientobilharzia in Clado-
grams 4 and 11 and is defined by Character 16 (derived

loss of Laurer's canal).

 

 

 

 

182

In Cladograms 5 and 12, New World Macrobilharzia
group with the North American mammalian schistosomes on
an optimized branch defined by the large number of eggs
in the uterus (Character 12). Cladograms 6 and 13 share
a basal multichotomy and include a number of components
absent in Cladograms 2 and 9, such as Components 16 and
20.

Finally, Cladograms 7 and 14 base three Inajor nodes
on components described by character states referring
to testes number (Character States 241, 242 and 243).
This results in the early branching off of taxa with low
testes numbers: Schistosoma, Schistosomatium and Austro—
bilharzia.

All of the 12 non—preferred cladograms include compo-
nents that increase the number of postulated evolutionary
changes. Interestingly, of the branching diagrams proposed
by this analysis, the arrangement of taxa in the preferred
cladograms differs least from currently proposed subfamilial
relationships within the Schistosomatidae (Farley, 1971).
Nonetheless, the results of this analysis bring into focus
questions concerning host relationships of taxa previously
assigned to the subfamily Schistosomatinae. Although

avian and mammalian genera appear in the same clade, they

cluster into discrete units which reflect host class affili—

ation.

 

 

 

 

 

183

Systematic Implications

The cladistic relationships proposed herein are in
no way the final word on schistosome phylogeny and I expect
further data to refine and perhaps redefine the phylogenetic
structure I propose. This is particularly true since
relatively few characters or character states define many
branches of my preferred cladogram. Although I would
like to have more data available before attempting a com-
plete taxonomic revision of the family, two reinterpre-
tations of current systematics are worth mentioning.

I see no utility in retaining the subfamilial rankings
of Dendritobilharziinae and Gigantobilharziinae. Cladistic
analysis clearly indicates inclusion of the genera contained
in these two subfamilies in the subfamily Bilharziellinae.
Accordingly, in subsequent discussion I will consider
the subfamily Bilharziellinae to contain the genera Bilhar-
ziella, Trichobilharzia, G'gantobilharzia and Dendrito—
bilharzia. Further study is needed to clarify relationships
within the Bilharziellinae, particularly with regard to
the genera Gigantobilharzia and Dendritobilharzia which
share a unique terminal opening of the female genital
pore. Here and throughout the study, more complete infor-
mation on cercarial characteristics will help with the
differentiations I wish to make.

The subfamily Schistosomatinae is probably a poly—
phyletic grouping as currently defined. A conservative

change would involve removing the genus Macrobilharzia

 

184

to its own subfamily, which could be named the Macrobil-
harziinae. The genus is poorly understood and it retains
many primitive character states. I am currently unable
to clarify its relationship to the other two schistosome
subfamilies. For the present time, I have chosen to retain
the mammalian genera Schistosomatium, Heterobilharzia
and Bivitellobilharzia in the subfamily Schistosomatinae.
Further research may suggest their removal based on a
number of derived character states which differentiate
them from the other members of the subfamily.

I have already discussed the problems with current
definitions of the genera Macrobilharzia and Sinobilharzia.
These genera await further data which will clarify their
status. In Appendix 1, I have listed the genera recognized
in this study and most of the reports describing ”species”
within each genus. In a few cases, I have noted generic
synonomies which do not exist in the current literature,
but for the most part I have not commented on the validity
of described species. An understanding of the status
of the species in all genera will only be achieved following
considerable study of morphological variability and genetic
species boundaries. Although some progress has been made,
the area of speciation biology in schistosomes offers

many unanswered and intriguing questions for the future

Of parasite systematics.

 

 

 

185

Phylogenetic Results

Phylogenetic analysis of character state evolution
in the Schistosomatidae provided me with two preferred
cladograms that I feel represent the best estimate of
schistosome phylogeny. The two preferred cladograms differ
in the placement of the genus Dendritobilharzia. Largely
for ease of interpretation, in the following analysis
I will focus on the cladogram that places Dendritobilharzia
and Gigantobilharzia as the terminal node in the Bilharziel-
linae (Figure 30). I tend to favor this placement based
on the peculiar derived location of the female genital
pore but switching Dendritobilharzia and Trichobilharzia
does not alter the results of host and geographic analyses.

To examine hypotheses of cospeciation and host shift-
ing, I examined patterns of congruence between the distri-
butions of hosts superimposed upon parasite phylogenies
in comparison with independently derived theories of host
relationships. If coevolution has been a theme in schisto-
some evolution, then some degree of congruence would be
expected between parasite and host phylogenies, either
at the intermediate or definitive host level, or both.
Incongruities between host and parasite phylogenies tend
to deny hypotheses of coevolution, and suggest rather

that host shifting has been a factor during parasite evolu-

tion.

 

 

 

 

 

 

 

 

Figure 30. Preferred. cladogranl depicting' the phylogenetic

relationships of the Schistosomatidae.

 

 

187

 

 

 

 

 

 

188

Vertebrate Host Relationships

Figure 31 depicts the preferred cladogram of the
phylogenetic relationship of the schistosomes with avian
and mammalian host taxa assigned to the terminal nodes.
Host taxa are usually presented at the ordinal level,
since in most cases a number of families and/or genera
are inhabited. Two of the three basal branches, the sub-
families Macrobilharziinae and Bilharziellinae, contain
members which occur only in avian hosts. The third branch
includes both birds and mammals as hosts. Of the two
lineages in the third branch, one occurs only in mammals
while the other is primitively found in birds and second-
arily occurs in mammals.

From the outset, it appeared likely that host shifting
was involved at some time in the evolutionary history
of the schistosomes because the two major definitive host
groups, birds and mammals, are only distantly related,
having been separated from one another by several reptilian
lineages during their phylogenetic history. To determine
which taxon is likely to be the primitive host group for
the lineage, host taxa can be treated in a manner analogous
to conflicting character states (Mickevich, 1981). Working
at the level of vertebrate class, I applied a Farris optimi-
zation procedure in which ancestral hosts were assigned
to interior nodes. Results of the optimization procedure
indicate that birds are the primitive hosts of the schisto-

somes and that a minimum of two shifts into mammals have

 

 

 

 

 

 

 

 

189

Figure 31. Vertebrate host distributions assigned to
the preferred cladogram depicting relationships
of the Schistosomatidae. References documenting

host distributions are presented in Appendix

I.

190

moEco=_..uo.ozo

moetorccoofia
$582..me
wrong. to $0. 5;. ”no. a. £38 x
negozcuugon. $652524
noggin—v.65
maststoenfio 3E3332<
3.5056230
36.81024
mics—__suaazu
353.395
23032.24
83:35 3.58.5»: 32889.... 90358 Essex 352.383; 35268668 x onazuwnaaon; mesrchudx
2.ng / \ magazcuogoa x 3.52.8.4 x
agate x 35356220 35536055 x $6555.:on
22:08:. x 82:on N
36.523me x
25032.3 $552.85..

 

gned t0
re1ationships

 

 

 

191

occurred during the evolutionary history of the Schisto-
somatidae. At the vertebrate class level, therefore,
host shifts are suggested but questions of cospeciation
must be addressed within each vertebrate lineage.

Optimization indicated that three avian orders or
their ancestors served as hosts for the three lineages
of the preferred cladogram. These three avian taxa are
the Anseriformes (ducks and geese) which host members
of the Bilharziellinae, the Pelecaniformes (cormorants
and anhingas) which host the genus Macrobilharzia, and
the Charadriiformes (gulls, terns and shorebirds) which
host avian taxa assigned to the Schistosomatinae. Optimi-
zation does not designate a single basal taxon but rather
results in a union of the three avian orders.

Although the interrelationships between orders of
birds are at best poorly known (Cracraft, 1981), the orders
portrayed here are not considered to have shared a recent
phylogenetic history. It is impossible to rule out host
shifts between avian host taxa, but a parsimonious interpre-
tation of definitive host relationships suggests that
the common ancestor of the anseriforms, charadriiforms
and pelecaniforms hosted the ancestor‘ of modern schisto-
somes. Relatively broad host specificity may explain
reports of minor hosts (birds reported in less than ten
percent of accounts) among aquatic birds; until more
data are available, I cannot determine if minor host taxa

represent shifts or chance infections.

 

 

 

 

192

The pattern of mammalian distribution portrayed in
Figure 31 suggests a minimum of two independent shifts
into mammals. One invasion gave rise to an ancestor of
the genera Bivitellobilharzia, Heterobilharzia and Schisto-
somatium, while an ancestor of Schistosoma and Oriento-
bilharzia invaded xnammals at a later date. Unlike avian
blood flukes, mammalian genera show a high degree of speci-
ficity at the level of host order, with the exception
of the genus Schistosoma. Even so, this specificity does
not readily appear to reflect phylogenetic congruence
with mammalian host evolution (see Simpson, 1945, and
McKenna, 1975, for reviews of mammalian phylogeny). The
five or six Inajor Inammalian orders involved share common
ancestors very early in the evolution of placental mammals,
perhaps suggesting an early origin of the shifts which
gave rise to mammalian lineages. Much more recent host
shifts at the level of individual species or species groups
probably account for the distribution of some Schistosoma

in primates, rodents and carnivores.

Intermediate Host Relationships

Although missing data in the form of unreported host
taxa may affect the analysis of definitive host relation-
ships, its impact upon the understanding of intermediate
host biology is more readily apparent. Intermediate life
history stages have not been described for the genera

Dendritobilharzia, Macrobilharzia (Old and New World forms)

 

 

 

 

 

 

193

and Sinobilharzia. Figure 32 depicts the known intermediate
host distributions of the Schistosomatidae. Members of
three gastropod subclasses serve as hosts: Opisthobranchia,
Prosobranchia and Pulmonata. The latter two taxa are
by far the most common, and the majority of blood flukes
are found in the pulmonate families Planorbidae and Lymnaei-
dae.

Farris optimization (Figure 32) indicates that the
pulmonates are the primitive hosts of the schistosomes.
At least one host shift has occurred into opistobranchs
by some members of the genus Gigantobilharzia: two
researchers, one in Europe and one in North America, have
recovered Gigantobilharzia cercariae from marine
opistobranchs (Donges, 1964, and Leigh, 1955). The genera
Austrobilharzia and Ornithobilharzia occur only in marine
prosobranch gastropods (Farley, 1971). While the majority
of flukes in the genus Schistosoma occur in planorbid
pulmonates, two closely related species, S. japonicum
and §, mekongi, are found in freshwater prosobranchs.
Figure 32 also depicts a shift back into pulmonates by
the genera Schistosoma and Orientobilharzia.

All schistosomes recorded from pulmonates are found
in one or more of three closely related families: Physidae,

Planorbidae, and Lymnaeidae. Certain genera (e.g., Giganto-

bilharzia, Trichobilharzia and Schistosoma) are found
in more than one pulmonate family. Unless host shifting

among pulmonate families has occurred after the

 

 

."'

 

 

 

 

Figure 32.

194

Intermediate host distributions assigned to
the preferred cladogram depicting relationships
of the Schistosomatidae. References documenting

host distributions are presented in Appendix

I.

|\\I‘|l1l|u||ll]
7 .

195

253.. .o 30. 8.: 32 5 £380 x

EucoEan.

25:05.88...

02205085

2286.3.

oEoEESoix 8385.3
3209..ij Basis...

009.2555
055.53.

255.5035 « Juu_wc..\E3

EacoE3m

relationships

\

Eocofisd

a.

”8555...

822.5 x

EocoESaooEmccED

EocoESm

aficofifia

0.08.3.5

2655.55

332:3

oBcoESa
o_;uco£o..=m_no x

 

 

 

196

establishment of those families, the observed distribution
suggests that schistosome genera predate diversification
of the pulmonates. It seems to me more likely that some
host shifting has occurred since the origin of major
molluscan lineages; the Jurassic origin of the three
pulmonate families (Zilch, 1959) significantly predates
the origin of eutherian lineages of mammalian definitive
hosts (Lillegraven et al., 1979).

With the exception of Davis' monography on the
pomatiopsid snails (Davis, 1979), there are no other
published discussions of gastropod phylogeny which utilize
cladistic methodology. Figure 33, assembled from
information in Fretter and Peake (1978), depicts current
thinking on the relationships among the gastropod taxa
of interest in this study. This phylogeny offers little
useful information for my analysis, since schistosome
evolution seems primarily associated with pulmonate
gastropods, accompanied by an amount of shifting among
host taxa (Figure 32) and detailed phylogenies of the

pulmonata are unavailable.

Biogeographic Relationships

Cladistic analysis of organismal distributions has
become an increasingly important technique in reconstructing
phylogenetic history. This methodology is based primarily
On an analysis of organismal distribution relative to

continental drift theory. Brooks (1978) demonstrated

 

 

 

 

 

 

 

197

Figure 33. Probable phylogenetic relationships of the
molluscan intermediate hosts of the Schisto-

somatidae (adapted from Fretter and Peake,

1978).

198

the

of

ships

the Schisto’

and Peake»

Pulmonata

 

 

 

the utility of this methodology in a parasitological context
in his study of the cestode fauna of crocodilians. I
examined the geographic distributions of the schistosomes
in an attempt to understand the potential relationship
between host dispersal and Vicariant pattern.

In order to examine biogeographic pattern, I assigned
geographic locations to the terminal nodes of the preferred
cladogram (Figure 34) in a manner analogous to that used
in the host analyses. The geographic locations consisted
of continental or subcontinental land masses. I compared
the cladogram of schistosome distributions to a recently
proposed cladogram (Figure 35) depicting the sequence
and relationship between land masses during continental
breakup (Humphries, 1981). There is no apparent congruence
between the two cladograms or between portions of them.

Farris optimization produces a basal grouping of
land masses (Figure 34) that bears no resemblance with
proposed historical land masses with the exception of
Pangea. In fact, a Pangean origin of the schistosomes
is consistent with the ages of host inhabitation implied
in the previous analyses. Ancestral schistosomes appear
to have inhabited the common ancestors of modern host
orders; host shifts from birds to mammals apparently occur—
red before the origin of modern orders. Conservative
estimates place the origin of common ancestral hosts prior
to or at the beginning of the breakup of Gondwanaland

(Lillegraven et al., 1979, Van Tyne and Berger, 1976).

 

 

 

 

200

Figure 34. Farris optimization of schistosome geographic
distributions. References documenting geo-

graphic distributions are presented in Appendix

1.

some geographic
cumentiflg geo-

Ited in Appendix

 

IN
IN AS
AS AF IN
AF EU AS
EU SA EU
NA NA NA
AS,AF,EU,
IN,NA
IN,AS,EU,NA
IN,EU

PH IN
AS HA AS
IN EU AU IN AF
AF SA IN IN SA AS AS EU
EU NA AF NA NA AF AS NA N EU SA
AS.EU,
IN
AS
NA, SA. NA. IN, AF AS
IN, AF
AS
NA,AF,IN.AS
AF = Africa IN = India
IN,EU,NA, AS = Asia . NA= North America
SA,AF,AS AU = Australia PH: Phippines

EU = Europe SA = Sauth America
HA = Hawaii

 

 

 

 

 

 

 

 

 

 

202

Figure 35. Proposed sequence of relationships between
land masses during continental breakup (adapted

from Humphries, 1981).

 

North . , New East , , WesI . South,
America Eurasm Africa Zealand Antarctica Australia Antarctica America

India

nshiPS heme“

)reakup (adapted

 

   

 

 

 

 

204

It also seems likely that some geographic dispersal has
occurred via migratory birds since continental breakup.
This possibility is suggested by the widespread distribution
of avian schistosomes relative to those genera inhabiting
mammals. With the exception of Schistosoma, mammalian
genera are restricted to either the Old or New World,

while several avian genera are nearly circumpolar in distri-

bution.

Evolution of the Schistosomatidae

A Synthesis

In this section, I will present an overview of schisto-
some evolution set in a context of phylogenetic and evolu-
tionary history. I will attempt to unite the information
generated in this study intc> a cohesive framework chron-
icling patterns of evolutionary change.

The blood flukes along with other families in the
trematode suborder Strigeata form :1 well defined lineage
based primarily upon cercarial morphology. Phylogenetic
analysis indicates that a three host life cycle, involving
one invertebrate and two vertebrate hosts may be the
primitive developmental pattern for the group. Loss of
one vertebrate host may have subsequently occurred in

each of the three blood fluke families.
The primitive invertebrate intermediate hosts of
the lineage appear to be members of the freshwater molluscan

class Pulmonata. Clinostomes and all strigeiform families

 

 

 

 

205

have been reported only from pulmonates. The three families
of blood inhabiting flukes all contain some members which
develop in marine molluscs other than pulmonates, and
in sanguinicolids, development probably occurs in annelid
worms as well. I suggest that the occurrence of non-
pulmonate intermediate hosts in the life cycles of the
three blood fluke families represents independent host
shifts by certain members of each family. These host

shifts are correlated with and may have accompanied invasion

of marine habitats by formerly freshwater parasites. I
will consider this point in more detail in my discussion
of the family Schistosomatidae.

Within the Strigeata, the clinostomes and blood flukes
appear to form the sister group of the strigeiform families.
Although the spirorchids and schistosomes clearly form
a monophyletic lineage, the relationships of the clinostomes
and sanguinicolids to each other and to the tetrapod blood
flukes is ambiguous. I feel that current evidence does
not support the classical interpretation of monophyly
of the blood flukes. Characters which unite the blood
flukes involve morphological loss and certain reports
indicate the major characters lost (absence of pharynx
and redial stage) may be present in some members of the
Sanguinicolidae. In contrast, a complex cercarial body

form, the lophocercaria allies clinostomes and sanguini-

colids.

 

 

, -.

 

 

 

 

I favor an interpretation in which the clinostomes
and sanguinicolids form a monophyletic sister group of
the tetrapod blood flukes. All four families were derived
from :1 common ancestor which utilized a three host life
cycle. Therefore, this suggests that clinostomes retained
the primitive developmental pattern while at least two
independent losses of a vertebrate host stage occurred

in the blood flukes.

Presuming that the sanguinicolids and clinostomes
shared a recent common ancestor, I suggest that it was
the second or definitive vertebrate host which was lost
by the sanguinicolids. Rather than encysting upon the
surface of the intermediate piscean host, perhaps ancestors
of the sanguinicolids penetrated into the host directly
and completed their life cycle. This interpretation differs
markedly from the classical view which sees development
in the host circulatory system as a major reason for uniting
the blood fluke families in a single superfamily.

The origin of direct penetrating by the spirorchid-
schistosome ancestor most likely predated the diversifi-
cation of the common reptilian ancestor which gave rise
to modern host groups. In this case, the encysting phase
associated with a vertebrate intermediate host was elimi—

nated and penetration of the final reptilian host occurred
directly.
At least two blood fluke lineages appear to have

coevolved with different derivitives of early reptiles.

 

 

 

207

One lineage, the Spirorchidae, was isolated in turtles
or their ancestors while the schistosomes developed in
the reptilian ancestors of modern birds. At some point
prior to the diversification of early avian lineages,
the ancestors of modern schistosomes added a major evolu-
tionary novelty in the form of dioecy, a synapomorphic
character which defines the family.

The family Schistosomatidae may have had its orign
in Pangea prior to continental breakup, as suggested by
the current world—wide distribution of the family. Modern
schistosomes may have shared a common ancestral host found
in the avian lineage which gave rise to anseriform,
charadriform and perhaps pelecaniform birds. Of the three
recent schistosome lineages, two are found only in birds,
while one utilizes both birds and mammals as hosts. Within
the Schistosomatinae, cladistic analysis suggests that
distribution in mammals is the result of at least two
separate host shifts during the evolutionary history of
the schistosomes.

Character analysis clearly defines genera of the
subfamily Bilharziellinae as a monophyletic group. Anseri—
form birds (ducks, geese and swans) or their ancestors
appear to be the primitive hosts of the Bilharziellinae.
Although the optimization procedure labeled one avian
order as primtively hosting the subfamily, within most
genera host specificity seems somewhat loose and several

other avian taxa appear as minor hosts in distributional

 

 

 

208

records (Figure 31). It appears that an ecological rather
than systematic theme relates the different minor host
taxa, that is, hosts recorded from less than ten percent
of reports for a given genus. This theme is apparent
in the observation that, to a. greater or lesser degree,
all of the avian hosts are found in aquatic habitats.
The non-aquatic passerines (perching birds) that are found
as minor hosts in the genera Trichobilharzia and Giganto-
bilharzia are the only exceptions to an aquatic trend.
The lack of relationship among the birds inhabited, as
well as the number of different host species infected
in certain families (e.g., Anatidae), imply that it is
largely the fact of living in an aquatic habitat that
determines which host species are infected. The underlying
ordinal trends suggest that certain avian groups may serve
as more suitable hosts than others, but the overall lack
of specificity makes more definite statements difficult
at the present time.

Limited distributional records indicate that passerine
hosts occur in the morphologically similar genera Giganto-
bilharzia and Trichobilharzia (Ito, 1960; Byrd, 1957;
Najim, 1950; and Brackett, 1942). I feel it is impossible
to determine whether these reports represent chance inva-
sions of bathing birds or whether they indicate a. close
relationship between the two genera. Following the argu—
ments used in determining character state distributions,

the fact that anatids, the primitive hosts of the lineages,

 

 

 

 

209

occur in both gggppppilpgpgig and gigappppilhgpgig suggests
independent invasions of passerines. Evolution from an
anseriform inhabiting ancestor may have been followed
by shifts into passerines by a few species in each genus.

Pulmonate gastropods of several families (Figure
32) serve as intermediate hosts for members of the
Bilharziellinae and at least one shift into opisthobranch
gastropods occurred in the genus Gigantobilharzia. I
suspect that a better understanding of species relationships
within schistosome genera will reveal more detailed patterns
of coevolution with different pulmonate families. Cur-
rently, incomplete life cycle data and poorly defined
species boundaries hamper this inquiry.

Distributional records for members of the subfamily
Macrobilharziinae are restricted to birds in the order
Pelecaniformes. Anhingas serve as hosts in the Americas,
while cormorants host Macrobilharzia in both Africa and
India. Formerly placed in separate families, cormorants
and anhingas are currently grouped in the family
Phalacrocoracidae (Cracraft, 1981). Distributional data
suggest an historical pattern of coevolution. Cormorants
and anhingas each occur on both sides of the Atlantic
Ocean. Present parasite distributions imply either isola-
tion in different divisions of the Phalacrocoracidae follow—
ing continental breakup or extinction in one host group
in each portion of the disjunct range. Further information

in the form of additional host records and more detailed

 

 

 

210

morphological study may aid in distinguishing between
these hypotheses.

Macrobilharzia retains more primitive morphological
characters than any other schistosome genus. Therefore,
it is interesting to note that the host order, Pelecani-
formes is considered by most workers to be primitive rela-
tive to the other major schistosome host groups, the Anseri-
formes and Charadriiformes (Cracraft, 1981). Thus host
distribuion of the Macrobilharziinae may conform to a
modified version of the Szidat Rule, in which parasites
retaining primitive traits inhabit host groups which also
possess primitive morphological attributes. In this case,
host data may suggest a resolution of the basal trichotomy
in the preferred cladogram of schistosome relationships
(Figure 31). Althouga I would like to see morphological
evidence to confirm this resolution, the Macrobilharziinae
may represent the earliest offshoot from a primitive
schistosome ancestor.

In the third schistosome subfamily, the Schistosoma-
tinae, charadriiform birds appear as primitive hosts and
at least two shifts of mammalian hosts have occurred in
the phylogenetic history of the group. One avian genus,
Austrobilharzia, is frequently reported from anseriform
as well as charadriiform birds. This observation suggests
an avenue for further study to determine whether inhabiting

different host orders reflects undetected morphological

differences within the genus.

 

 

 

211

Avian members of the subfamily Schistosomatinae develop
in prosobranch gastropods, while with two exceptions,
mammalian genera retain the primitive schistosome inter—
mediates, pulmonate gastropods. Shifts by avian taxa
into prosobranch snails may have accompanied or precipitated
invasion of a new habitat, the marine environment.

Within the Schistosomatinae, an early shift into
mammals may have occurred by the common ancestor of Schisto-
somatium, Heterobilharzia and Bivitellobilharzia. The
two genera of North American blood flukes, Schistosomatium
and Heterobilharzia are clearly more closely related to
one another than to other genera of schistosomes. Both
genera are primarily restricted to their rodent and carni-
vore hosts, respectively. Within host orders, however,
specificity breaks down and members of various families
serve as hosts. For example, microtine, peromyscine and
zapodid rodents host Schistosomatium, while raccoons,
feral dogs and bobcats are the commonly reported hosts
of Heterobilharzia (see references in Appendix 1). In
the case of North American schistosomes, host distribution
does not offer clear evidence of systematic affiliation
of the parasite taxa. The relationships of rodents to
other mammalian taxa remain unclear after years of study,
and no author has closely associated rodents with carnivores
(McKenna, 1975). The lack of clear relationships among
host taxa suggest a shift into mammals by the ancestors

Of North American blood flukes very early in placental

 

 

 

212

history. Alternatively a shift by the ancestors of one
genus may have given rise to a new parasite taxon in a
different host group.

Host relationships of the Afro-Indian genus Bivitello-
bilharzia, as with most aspects of the genus, are prob-
lematic. The genus is known on the basis of two descrip—
tions, one from a wild caught African elephant (Loxodonta
africana) on display in a German zoo and one very vague
account from zul elephant (Elephas maximus) in India. If,
as I suggested earlier, the host shift which established
the mammalian lineage containing Bivitellobilharzia took
place early in placental evolution then the relationship
of the genus to the North American blood flukes is not
unfeasible. All three host orders shared a common ancestor
at some point in their history, but our current knowledge
of this phase of mammalian history is fragmentary at best.
Alternatively, and perhaps more feasibly, a host shift
into the ancestor of modern elephants may account for
the distribution seen today.

A second shift into mammals occurred with the ancestors
of the genera Orientobilharzia and Schistosoma. These
two genera are quite similar in morphology, differing
only in the number of testes and the absence of a seminal
receptacle in Orientobilharzia. The genus Orientobilharzia
is Eurasian in distribution and found entirely in bovids.
The genus Schistosoma appears quite diverse relative to

other members of the family. There are approximately

 

 

 

213

17 described species of Schistosoma (Davis, 1980), more
than in any other schistosome genus except Trichobilharzia.
This relatively large number of species may simply reflect
the great amount of attention the genus has received due
to the fact that some species parasitize humans. The
possibility exists that other genera may have species
which have not been reported.

Species diversity in Schistosoma appears to be related
to variations in infectivity to different host taxa (Wright,
1960). Specificity and strain variation seems most pro-
nounced at the intermediate host level; in contrast, the
genus and certain species show the greatest diversity
of definitive host species to be found in the family.
Schistosoma mansoni, for example, has been reported natur—
ally from humans, non—human primates, insectivores and
many rodents.

Intermediate host relationships of the genus Schisto-
§Qma have received considerable attention. The genus
occurs primarily in planorbid pulmonates, but at least
two species, ‘S. japonicum and S, mekongi, have shifted
into prosobranch snails. It was, in fact, the occurrence
Of S. mekongi in a previously undetected prosobranch snail
that alerted researchers to the separate taxonomic status
of this parasite (Bruce and Santasiri, 1980). Detailed
studies subsequently demonstrated morphological and bio-

chemical differences that resulted in specific differen-

tiation.

 

 

214

The genus Schistosoma is widespread in distribution across

Asia, southern Europe, Africa and South America. The
range <xf one species in particular has drawn a consider—
able amount of attention: S, mansoni is found in both
Africa and South America, where it develops in closely
related snails of the genus Biomphalaria. On both conti-
nents, a variety of mammalian hosts is infected, with
various rodents and man reported most frequently (Kuntz,
1955). Studies show that Old and New World forms are
interfertile, although varying degrees of intermediate
host specificity exist in different populations, both
within and between continents (Kuntz, 1955).

The transoceanic distribution of S, mansoni has classi—
cally been interpreted as the result of an introduction
from Africa into South America via the slave trade (Cheng,
1973). Recently, Davis (1980) presented an alternative
Vicariant explanation. Based on his studies of gastropod
biogeography, Davis proposed that populations of S. mansoni
inhabiting snails of the planorbid genus Biomphalaria
were isolated on opposite sides of the Atlantic following
the breakup of Gondwanaland. Separation of Africa and
South America was complete by the late Cretaceous, between
70 and 100 M.Y.B.P. (Lillegraven et al., 1979). By the
early Cretaceous, members of many modern genera of planor-
bids and lymnaeids were present (Zilch, 1959) and a vicar-
iant explanation for the distribution of Biomphalaria

seems feasible. The genus Biomphalaria has speciated

 

 

 

215

extensively on both sides of the Atlantic since its original
separation (Davis, 1980). The situation with IS. pgupflyyi
requires more careful consideration. Davis is proposing
that not only the genus Schistosoma but also the species
S. mansoni have been separated by the Atlantic Ocean for
at least 70 million years. S, mansoni has failed to undergo
speciation in the Neotropics and it remains interfertile
with African forms after all this time. This would be
a remarkable example of stasis.

Although the intermediate hosts of _S. mansoni have
changed little since the time of continental breakup,
mammalian hosts have changed radically. At the time of
the separation. of South America and Africa, none of the
modern placental orders which serve as hosts had yet evolved
(Lillegraven. et al., 1979). Indeed, the placentals which
were isolated in South America gave rise to groups which
later became extinct (e.g., Meridiungulata); modern placen-
tal groups (e.g., rodents, carnivores, etc.) are presumed
to have reached South America. at a later date. In the
Neotropics, S. mansoni is found in several native rodents
as well as in introduced exotics such as Rattus and ngg
(Kuntz, 1955), but it is not found in remnants of the
archaic mammalian fauna, the didelphoid marsupials. It
is clear therefore that if .§- mansoni has been isolated
in South America since the breakup of Gondwanaland, shifts

into modern placental mammals are necessary to explain

current host distribution patterns. Much more recent

 

 

 

216

shifts would have to be postulated to explain the occurrence
of S. mansoni in Rattus and Spmg, two genera with relatively
recent histories in the Neotropics.

Given the information available, I think it is likely
that S. mansoni was introduced to South America following
the arrival of Old World people. The presence of suitable
snail intermediates which appear to show a Vicariant distri-
bution pattern allowed for successful colonization in
the New World. Perhaps the most compelling evidence in
support of this hypothesis are the observations that Old
and New World populations of S. mansoni are interfertile,
and while species diversity in Africa is high, the genus
has not undergone speciation during the proposed millions
of years of isolation in South America. It also seems
likely that the current distributions of human infecting
Schistosoma reflect the range limits of suitable inter—
mediate hosts, since human dispersal has, in essence,
been cosmopolitan.

Analysis geographic patterns indicates that avian
schistosomes tend to be quite widely distributed (Figure
34). This trend appears to be correlated with the greater
mobility of birds relative to mammals, and many avian
schistosomes inhabit migratory waterfowl, gulls and shore-
birds. Also, at least at the generic level, most avian
schistosomes inhabit, at least facultatively, a wide range
Of water dwelling birds, further increasing the probability

of widespread distribution. The wide distributions of

 

 

 

 

217

avian blood flukes suggest that intermediate host specifi-
city may be relatively weak, thus allowing cycles to 'be
established in new habitats when introduced by Inigrating
birds. Unfortunately, the intermediate host biology of
most avian schistosomes is quite poorly known, so this
suggestion awaits further study.

In contrast with avian forms, mammalian schistosomes
tend to be more limited in their geographic (and host
taxonomic) distributions. This correlates with the more
limited mobility of their hosts, and may also relate to
narrower definitive and/or intermediate host specificity.

In summary, the available information suggests that
the schistosomes are an ancient group which have been
associated with modern vertebrate lineages throughout
their evolutionary history. The vagaries of extinction
and host shifting have obscured patterns of ancient asso-
ciation. In addition, the mobility of avian hosts as
well as the broad specificity displayed by avian forms
may contribute to the lack of a coherent biogeographic
pattern. Wright (1960) suggested that intermediate host
speciation patterns constrain schistosome evolution, but
much more information is required before evaluating the
impact of intermediate hOSts on schistosome biogeography.
If schistosomes are, in fact, an ancient group, then it
appears that intermediate host relationships have remained
stable at least at the family and often genus level while

major vertebrate lineages have evolved and gone extinct.

 

 

 

 

 

218

This is suggested by the observation that recent pulmonate
host genera are identifiable in the fossil record prior

to the orign of nearly all modern vertebrate host taxa.

Biological Considerations

The observation that parasite species preferentially
infected certain host taxa suggested to early researchers
that host-parasite relationships reflect a history of
coevolution. Biochemical and physiological adaptation
of the parasite and its host constitute one aspect of
coevolution but it is the historical context of the host-
parasite interaction which I examined in this study.
Phylogenetic analysis utilizing cladistic methodology
provided a set of techniques which I used to test the
hypothesis that the phylogeny of the Schistosomatidae
reflects a history of coevolution with their vertebrate
and invertebrate hosts.

Phylogenetic analyses did, in fact, reveal an overall
pattern of coevolution at both levels of host involvement.
This coevolutionary pattern appears to be an ancient one
which was established prior to the origin of modern host
groups. The extreme age of the host-parasite relationship
is suggested by' the observation that parasite specificity
is most pronounced at the level of host order and that
within each vertebrate class, the orders which serve as
hosts are only distantly related. Although the lack of

a close relationship among host orders could also be

 

 

 

 

219

explained by postulating recent host shifts, until further
evidence to the contrary is available, coevolution may
be a more parsimonious explanation.

While coevolution has been the dominant theme in
the phylogenetic history of the schistosomatidae, host
shifting also played a prominent role in structuring current
host distributions. My analysis indicated that shifts
from birds into mammals were involved in establishing
at least two schistosome lineages. At the intermediate
host level, the ancestors of one avian lineage shifted
into marine prosobranch snails, and several species in
genera which generally develop in pulmonates shifted to
either prosobranch or opisthobranch gastropods.

Further study at the species level may reveal addi-
tional evidence of host shifting. Minor host taxa. may,
in fact, represent relatively recent host shifts by partic-
ular species or species groups, but more adequate sampling
along with complete life history data is needed to determine
whether host shifts have actually occurred.

The results of my research differ from other recently
published cladistic analyses of parasite phylogenies in
that host shifting appears to have played a significant
role in determining present relationships. Studies by
Brooks and his coworkers (summarized in Mitter and Brooks,
1983) describe complex patterns of coevolution, generally
focusing on species relationships within particular parasite

genera. I suspect that the difference in the frequency

 

 

 

 

 

 

220

of reported host shifts lies in the taxonomic level at
which the various studies were conducted. By analyzing
generic relationships, I maximized the chances of detecting
the effects of host shifting which occurred over an extended
period of evolutionary history. The studies by Brooks
and his coworkers focused on speciose parasite taxa which
display a narrow range of host specificity; thus groups
in which the likelihood of finding a coevolutionary pattern
is higher.
My research suggests that host shifts have played
a role in speciation and lineage formation during the
evolution of the schistosomes. This observation raises
questions concerning the interrelationships of host specifi-
city, host shifts and speciation mechanisms. At the verte—
brate host level, many schistosomes infect and successfully
reproduce in novel species (Bruce et al., 1961), but it
appears that relatively few infections of new host taxa
have led to speciation. As the growing body of literature
on parasite ecology indicates, complex behavioral and
biochemical factors function to maintain the intricate
balance which characterizes most host-parasite systems
(Price, 1980; Holmes, 1983). Davis (1980) noted that
trematode biology necessitates genetic adaptation to both
snail and vertebrate hosts within an environment suitable
for this three-way interaction. To understand the role
of host shifts in parasite speciation, research will have

to focus on the factors which change the balance between

 

 

 

 

221

host and parasite, thus providing the isolating mechanism
necessary to initiate speciation. Mechanisms may involve
shifts in the timing of reproduction in the altered immune
environment of a novel host or simply a transfer of the
system by the host into a different ecological habitat
(e.g., marine life zone) and away from the main parasite
population. Also, a change in the site of development
within a host species may sufficiently alter the course
of a life cycle to initiate isolation. For example, the
majority of schistosomes mature in the portal system of
their vertebrate hosts. In contrast, within the Bilharziel-
linae, the morphologically aberrant genus Dendritobilharzia
reproduces in the major anteries that leave the heart
of its anseriform hosts (Ulmer and Vande Vusse, 1976,
Vande Vusse, 1980). Novel behavioral patterns that precipi-
tate entering a new environment within a host individual
may be accompanied by shifts in the timing of reproduction
which temporally isolate a parasite from its surrounding
population. Questions concerning isolating mechanism
are addressed most appropriately at the species level
where fine scale differences in host distribution patterns
can be used to examine relatively recent speciation events.
With its complexities of host distribution at both the
intermediate and definitive host level, the genus Schisto-
§Qma_ suggests a starting point for such detailed studies

of speciation.

 

 

222

The question of the role of host specificity in the
evolution of the schistosomes has received considerable
attention at the intermediate host level. Wright (1960)
suggested that schistosome evolution was dependent upon
prior evolution of the snail host. He based this statement
on his observations of certain species in the genus Schisto-

soma. Wright's proposal may partially explain species

 

proliferation in the genus Schistosoma, but Davis (1980)
pointed out that schistosomes have in no way cospeciated
to match the post—Cretaceous radiation seen in modern
gastropod lineages.

Different species and genera of blood flukes show
very different patterns of intermediate host specificity.
Species of the African Schistosoma complex are interfertile
but show a high degree of intermediate host specificity
(Kuntz, 1955). In contrast, the widespread monotypic
genus Schistosomatium is found in a number of species
of lymnaeid snails, and worms from a given local can suc-
cessfully infect several different gastropod intermediates
(Malek, 1977). Variations in host specificity may represent
different ecological strategies which have evolved in
response to environmental factors or intrinsic host attri-
butes such as immune reaction. Broad host specificity
may reduce the likelihood that a parasite will closely
track evolutionary subdivisions of changing intermediate
host lineages. The ability to infect many different host

species may play a crucial role in schistosome survival

 

223

under unstable environmental conditions, but it may not
provide the set of circumstances which favor cospeciation.
Resolution of species boundaries will be the necessary
first step in examining the role of host specificity and
parasite speciation.

To study the speciation process of trematodes in
more detail, we will need to obtain a much better under-
standing of the nature of morphological character state
change during evolution. Conventional wisdom suggests
that morphological simplification is a major feature of
parasite evolution (Baer, 1951). Simplification may char-
acterize certain parasite groups, particularly arthropods,
but it is difficult to posit large scale simplification
when faced with the bewildering diversity of cercarial
morphologies displayed by parasitic trematodes (Schell,
1970). It seems more likely that trematodes have added
complex life stages during evolution from relatively simple
free living platyhelminths.

It was clear from the beginning of my research that
homoplasy, loss and reversal were prominent themes in
the evolution of the schistosomes and of trematodes in
general. For example, male terminal genitalia of all
parasitic trematodes appear to be homologous. Cestodes
and both monogenetic and digenetic trematodes possess
a presumably homologous series of organs which include
a cirrus, prostate, seminal vesicle, etc. The presence

Or absence of one or more of these structures varies between

 

 

 

 

224

and within families in all three platyhelminth classes.
Because of this intrafamilial variation, I avoided using
most male genital characters in my family level analysis.
Within the Schistosomatidae, I attempted to discern patterns
Nof derived loss using careful outgroup comparison. It
was in part because of the apparent frequency of character
loss that I was inclined to favor cladograms defined by
evolutionary gains rather than losses.

The study of changes in developmental timing that
result in heterochronic evolution may offer some insight
into the pattern of loss and reversal displayed by trema-
todes. Paedomorphosis is one form of heterochrony that
has frequently been associated with discussions of parasite
evolution(DeBeer, 1958; Gould, 1977), but few examples
of heterochronic development have been documented. In
my study of the Schistosomatidae, certain character state
changes imply that the host shift by the ancestor of
Orientobilharzia and Schistosoma was accompanied by
heterochronic changes in morphology. Fink (1982) carefully
developed the argument that heterochrony can only be
detected in an organism once the phylogeny containing
that organism has been determined. One set of circumstances
which alert a researcher to the potential occurrence of
heterochrony is the appearance of apparently primitive
characters rather than the expected derived ones at an
anomalous place on the chosen cladogram, such as the

terminal nodes of a derived lineage. Reversals to the

 

 

 

225

primitive ovary condition (Character States 70 and 71)
define the branch containing the genera 'Schistosoma and
Orientobilharzia. Also, the primitive condition of reduced
testes number (Character State 241) is found terminally
in Schistosoma. Although the evidence is not overwhelming,
it does suggest that further inquiry into the nature of
heterochronic change in these mammalian genera is
appropriate.

Heterochrony may explain what appears to be the fre-
quent loss and reversal in genital characters displayed
in comparisons between trematode families and genera.
Although genes coding for morphological characters are
not expressed, they are still present in the genome and
may become functional later in a given. phylogeny (Kollar
and Fisher, 1980). In order to test this hypothesis,
large scale examination of the patterns of character state
change will have to be based on detailed studies of morpho-
logical development. I feel that developmental studies
are likely to be the key to understanding the complexity
of homoplasy in trematode evolution. In addition, resolu-
tion of homoplasic characters will generate additional
sources of data which will aid in producing reliable phylo-
genies.

In proposing the ”parasitological method” nearly
twenty years ago, Hennig (1966) acknowledged the value
Of parasite data in constructing host phylogenies. In

a converse manner, parasitologists have often relied almost

 

 

 

 

226

exclusively on host relationships when determining parasite
phylogenies. When carefully constructed phylogenies of
both hosts and parasites are derived independently of
one another, a process of reciprocal illumination may
clarify relationships at both levels and focus further
attention on problematic branch points. Congruence between
host and parasite phylogenies can form the basis for more
detailed studies of coevolution, while areas of incongruity
offer evolutionary biologists models for examining the

role of host shifts and sympatric speciation.

 

 

 

 

 

 

APPENDIX 1

 

 

 

 

 

 

APPENDIX 1

Specimens and species descriptions used in assembling

character state,

host

and geographic distributions for

genera of the family Schistosomatidae.

Schistosoma

mansoni

japonicum

curassoni

bovis

margrebowiei

haematobium

 

intercalatum

 

incognitum

rodhaini

hippopotami

edwardiense

Bivitellobilharzia

loxodontae

nairi (=Schistosoma g.)

Faust et al. (1934)
Tada (1928)

Brumpt (1931); Gretillat (1962,
1964)

Khalil (1924)
LeRoux (1933)
Schwetz (1951)
Schwetz (1951)

Sinha and Srivastava
(1956, 1960)

Fripp (1967); Saoud (1966)
Thurston (1963)

Thurston (1964)

Vogel and Minning (1940)

Mudaliar and Ramanujachary
(1945)

227

 

 

Heterobilharzia

americana

Orientobilharzia

turkestanicum
(=Ornithobilharzia t.)

bomfordi

harinasutai

dattai

(=Ornithobilharzia g.)
Schistosomatium

douthitti

228

U.S.N.H.C. #S 14532, 39420
Lee, H.-F. (1962); Price
(1943)

U.S.N.H.C. #S 45820, 40932
Dutt and Srivastava (1961);
MacHattie (1934); Skrjabin
(1913)

Montgomery (1906)
Kruatrachue et al. (1965)

Dutt and Srivastava (1952)

Tanabe (1923); Price (1929);
El-Gindy (1951)

Austrobilharzia (=Microbilharzia)

chapini

terrigalensis
(6A1_xariglandi§

and Microbilharzia
variglandis)

(=A manitobensis
and A. canadensis)

penneri

pricei
(=Ornithobilharzia p.)

bayensis

U.S.N.H.C. #27888

U.S.N.H.C. #46655

Johnston (1917); Chu
and Cutress (1954);
Stunkard and Hinchcliffe

(1952)
McLeod (1937)

U.S.N.H.C. #39070
Short and Holliman (1961)

U.S.N.H.C. #29718
Wetzel (1930)

Tabangui (1933)

 

229

Macrobilharzia (New World)

macrobilharzia U.S.N.H.C. #27887

Travassos (1922); Price
(1931); Kohn (1964)

Macrobiharzia (Old World) (=Ornithobilharzia)

baeri Fain (1955)
phalacrocoraxi Baugh (1963)
Ornithobilharzia

\

 

 

canaliculata Dutt and Srivastava (1961);
Macko (1961); Witenberg
and Lengy (1967); Travassos
(1942)
(#9. intermedia) Odhner (1912)
($9. filamenta McLeod (1940)
and Q. aviani)
Sinobilharzia
odhneri
(=Ornithobilharzia g. U.S.N.H.C. #s 60990, 601015

 

and Austrobilharzia Q.) Faust (1924)

 

Bilharziella

polonica U.S.N.H.C. #14732
Baer (1932); Price (1929);
Kowalewski (1886); Khalifa
(1972); Ohdner (1912); Szidat
(l929a,b)

indica (=Chinhuta i.) Mehra (1940); Lal (1937)

lali Baugh (1963)

 

*Trichobilharzia (=Pseudobilharziella)

brevis U.S.N.H.C. #60715
Bash (1966)

crecci Lui and Bai (1976)
(=Jilinobilharzia c.)

230
jeguitibaensis Rios Leite et a1. (1978)
**Gigantobilharzia
lawayi U.S.N.H.C. #s 44862, 44863
Brackett (1942); Farley
(1963)
huronensis U.S.N.H.C. #37334

Najim (1950, 1956)

huttoni U.S.N.H.C. #37473
Leigh (1955)

Dendritobilharzia
pulverulenta U.S.N.H.C. #s 46655, 75331
Ejsmont (1929); Cheatum
(1941); Ulmer and Vande
Vusse (1970); Vande Vusse
(1980)
asiaticus Mehra (1940)

*For additional species and references, see Farley (1971) and
Fain (1956).

**For additional species and references, see Farley (1971) and
Fain (1960).

 

 

 

 

’r

LIST OF REFERENCES i

 

 

 

 

 

LIST OF REFERENCES

Agarwal, S. M. 1959. Studies of the morphology, system-
atics and life history of Clinostomum giganticum
n. sp. (Trematoda: Clinostomatidae). Ind. J.
Helminthol. 11275—115.

Anderson, D. J. and R. M. Cable. 1950. Studies on the
life history of Linstowiella szidati (Anderson) (Trema—
toda: Strigeatoidea: Cyathocotylidae). J. Parasitol.
36:395-410.

Azimov,, D. A. 1970. A revision of the systematics of
the Schistosomatata Skryabin and Schulz, 1937. Zool.
Zh. 49:1126-1131.

Baer, J. G. 1932. Contribution a la faune helminthologique
de Suisse. Rev. Suisse Zool. 39:1-57.

Baer, J. G. 1951. Ecology of Animal Parasites. Univ.
Illinois Press, Urbana.

Basch, P. F. 1966. The life cycle of Trichobilharzia
brevis n. sp. an avian schistosome fron1 Malaya. Z.
Parasitenkd. 27:242—251.

Basch, P. F. 1981. Cultivation of Schistosoma mansoni
in vitro. II. Production of infertile eggs by worm
pairs cultured from cercariae. J. Parasitol. 67:186-
190.

Baugh, S. C. 1963. Contributions to our knowledge of
digenetic trematodes VI. Z. Parasitenkd. 22:303-315.

Berrie, A. D. 1960. The influence of various definitive

hosts on the development of Diplostomum phoxini (Stri-
geida, Trematoda). J. Helminthol. 34:205—210.

Bilharz, T. 1852. Fernere Beobachtungen uber das die
Pfortades des Menschen bewohnende Distomum haematobium
und sein 'Verhaltniss zu gewissen pathologischen Bil-
dungen, aus brieflichen Mittheilungen an Professor
v. Siebold vom 29 Marz 1852. Ztschr. f. wissensch.
Zool. Leipz. 4:72-76.

231

 

 

 

232

Blankespoor, H. D. 1974. Host induced variation in Plagi-
orchis noblei Park, 1936 (Plagiorchiidae: Trematoda).
Am. Midl. Nat. 92:415-433.

Brackett, S. 1942. Five new species of avian schistosomes
from Wisconsin and Michigan with the life cycle of

Gigantobilharzia gyrauli (Brackett, 1940). J. Para-
sitol. 28:25-42.

Brooks, D. R. 1978. Evolutionary history of the cestode
order Proteocephalidae. Syst. Zool. 27:312-323.

Brooks, D. R. 1980. Allopatric speciation and non-inter—
active parasite community structure. Syst. 2001.
30:192-203.

Brooks, D. R.; D. R. Glen and R. T. O'Grady. 1983. Phylo-
genetic analysis of the Digenea (Platyhelminthes:
Cercomeria) with comments on their adaptive radiation.
Unpublished manuscript.

Bruce, J. I.; L. M. Llewellyn and E. H. Sadun. 1961.
Susceptibility of wild mammals to infection by Schisto-
soma mansoni. J. Parasitol. 47:752-756.

Bruce, J. I. and S. Sornmani, eds. 1980. The Mekong

Schistosome. Malacological Review, Whitmore Lake,
Michigan.

Brumpt, E. 1931. Description et duex bilharzies de mammi-
féres africans, Schistosoma curassoni sp. inquir
et Schistosoma rodhaini n. sp. Ann. Parasitol.
9:325—328.

Brundin, L. 1966. Transantarctic relationships and their
significance, as evidenced by Chironomid midges.
Kungl. Svenska Velenskapsa-kademiens Handingar Fjarde

series, 1111-472.

Byrd, E. E. 1957. Bilharziella littlebi (Byrd, 1956),
a blood fluke (Digenea: Schistosomatidae: Bilharziel—
linae) from the chestnut-sided warbler, Dendroica

pennsylvanica (Linnaeus). J. Parasitol. 43:374-376.

Carlton, M. D. 1980. Phylogenetic relationships in Neoto-
mine—Peromyscine rodents (Muroidea) with a reappraisal
of the dichotomy within New World Cricetinae. Misc.
Pub. Mus. Zool. U. Mich. 157:1-146.

Cheatum, E. L. 1941. Dendritobilharzia anatinarum n.
sp., a blood fluke from the mallard. J. Parasitol.
27:165—170.

 

 

 

 

233

Cheng, T. C. 1970. Symbiosis. Pegasus Press, New York.

Cheng, T. C. 1973. General Parasitology. Academic Press,
New York.

Chu, G. W. T. C. and C. E. Cutress. 1954. Austrobilharzia

variglandis (Miller and Northrup, 1926) Penner, 1953,
(Trematoda: Schistosomatidae) in Hawaii with notes
on its biology. J. Parasitol. 40:517—524.

Cobbold, T. S. 1859. On some new forms of entozoa. Trans.
Linn. Soc. London 22:363-366.

Cort, W. W. 1917. Homologies of the excretory system
of the forked-tailed cercaria. J. Parasitol. 4:49-57.

Cort, W. W. 1921. The development of the Japanese blood

fluke, Schistosoma japonicum Katsurada, in its final
host. Amer. J. Hyg. 1:1-28.

Cracraft, J. 1981. Toward a, phylogenetic classification
of the recent birds of the word (class Aves). Auk
98:681-714.

Cracraft, J. 1983. Cladistic analysis and vicariance

biogeography. Amer. Sci. 71:273-281.

Croft, J. W. 1933. A description of Cercaria whitentoni,
n. sp. Trans. Amer. Micros. Soc. 52:259-266.

Davis, G. M. 1979. The origin and evolution of the gastro-
pod family Pomatiopsidae, with emphasis on the Mekong

River Triculinae. Monogr. Acad. Nat. Sci. Phila.
20:1—20.
Davis, G. M. 1980. Snail hosts of Asian Schistosoma

infecting man: evolution and coevolution. Pp. 195-238
In The Mekong Schistosome (J. I. Bruce and S. Sornmani,
eds.). Malacological Review, Whitmore Lake, Michigan.

DeBeer, G. R. 1958. Embryos and Ancestors. Oxford Univ.
Press, London.

Dienske, H. 1968. A survey of the metazoan parasites
of the rabbit-fish, Chimera monstrosa L. (Holocephali).
Netherlands J. Sea Res. 4:32—58.

Diesing, K. M. 1858. ”Revision der Myzhelminthen Abt.
Trematoden. Sber. Ost Akad. Wiss., Wein. 32:307—
390.

 

D5n

234

ges, J. 1964. Gigantobilharzia suebica n. sp. (Trema-
toda), ein Dermatitiserreger beim Menschen. Z. Para-

sitenkd. 24:65-75.

Dubois, G. 1938. Monographie des Strigeida (Trematoda).

Mem. Soc. Neuchét. Sc. Nat. 6:1—535.

Butt, S. C. and H. D. Srivastava. 1952. On the morphology

and life history of a new mammalian blood fluke --
Ornithobilharzia dattai n. sp. (preliminary report).
Parasitology 42:144-149.

Dutt, S. C. and H. D. Srivastava. 1961. A revision of

the genus Ornithobilharzia Odhner (1912) with the
creation of two genera Orientobiharzia Dutt and
Srivastava (1955) and Sinobiharzia Dutt and Srivastava
(1955) (Trematoda: Schistosomatidae). Ind. J. Helmin-

thol. 13:61-73.

Ejsmont, L. 1926. Morpholoische, systematische und ent—

wicklungsgeschichtliche Untersuchungen an Arten des

Genus Sanguinicola Plehn. Bull. Int. Acad. Pol.
Sci. Lett. C1. Sci. Math. Nat. 9—10:877-966.

Ejsmont, M. L. 1929. fiber zei Schistosomatidengattungen

der V6gel. Bull. Acad. IP01. Sc. et Lettre C1. Sc.
Math et Nat. S. B. Nat (II):389-403.

Eldredge, N. and J. Cracraft. 1980. Phylogenetic patterns

and the evoutionary process. Columbia Univ. Press,
New York.
El—Gindy, M. S. 1951. Post-cercarial development of

Fain,

Fain,

Fain,

Schistosomatium douthitti (Cort 1914) Price 1931
in mice, with special reference to the genital system
(Schistosomatidae-Trematoda). J. Morph. 89:151—177.

A. 1955. Un nouveau schistosome du Comoran au
Ruanda-Urundi (Congo Belge) -- Ornithobilharzia baeri

n. sp. Acta Tropica 12:357-360.

A. 1956. Les schistosomes d'oiseaux du genre Tricho—
bilharzia Skrjabin et Zakharow, 1920 au Ruanda-Urundi.

Rev. Zool. Bot. Afr. 54:147—178.

A. 1960. Nouveaux schistosomes d'oiseaux du genre

Gigantobilharzia Odhner. Ann. Parasit. Hum. Comp.
35:292—304.

Farley, J. 1963. A redescription of Gigantobilharzia

lawayi Brackett, 1942. J. Parasitol. 49:465-467.

 

 

 

 

235

Farley, J. 1971. A review of the family Schistosomatidae
excluding the genus Schistosoma from mammals. J.

Helminthol. 45:289-320.

Farr, M. M. and V. G. Blankemeyer. 1956. Trichobiharzia
brantae n. sp. (Trematoda: Schistosomatidae) from

the canada goose (Branta canadensis L.). J. Para—
sitol. 42 320—325.

Farris, J. S. 1970. Methods for computing Wagner trees.
Syst. Z001. 19:83-92.

Farris, J.S. 1983. The logical basis of phylogenetic
analysis. Pp. 7-36 In Advances in Cladistics, Volume

2. (N. I. Platnick and V. A. Funk, eds.). Columbia
Univ. Press, New York.

Faust, E. C. 1924. Notes on Ornithobilharzia odhneri
n. sp. from the asiatic curlew. J. Parasitol. 11:50-
55.

Faust, E. C., C. A. Jones and W. A. Hoffman. 1934.
Studies on schistosomiasis mansoni in Puerto Rico.
III. Biological Studies. The mammalian phase of
the life cycle. P. R. J. Pub. Health Trop. Med.
10:133—196.

Felsenstein, J. 1978. Cases in which parsimony or compati-
bility methods will be positively misleading. Syst.
Zool. 27:401—410.

Felsenstein, J. 1979. Alternative methods of phylogenetic

inference and their interrelationship. Syst. Zool.
28:49-62.
Felsenstein, J. 1982. Numerical methods for inferring

evolutionary trees. Quar. Rev. Biol. 57:379—404.

Fink, W. L. 1982. The conceptual relationship between
ontogeny and phylogeny. Paleobiology 8:254—264.

Freeman, R. S. 1973. Ontogeny of cestodes and its bearing
on their phylogeny and systematics. Adv. Parasitol.
2:481—557.

Fretter, V. and J. Peake, eds. 1978. Pulmonates, Vol.
2A: Systematics, evolution and ecology. Academic
Press, London.

Fripp, P. J. 1967. On the morphology of Schistosoma
rodhaini (Trematoda, Digenia, Schistosomatidae).

J. Zool. Lond. 151:433-452.

 

 

 

 

236

Fuhrmann, O. 1928. Trematoda .ln Handbuch der Zoologie
(W. Kukenthal and T. Krumbach, eds). Berlin.

Funk, V. A. and D. R. Brooks, eds. 1981. Advances in
Cladistics, Volume 1. N. Y. Botanical Garden, Bronx.

Goodchild, G. C. and V. L. Martin. 1969. Speciation
in Spirorchis (Trematoda: Spirorchiidae) infecting
the painted turtle Chrysemys picta. J. Parasitol.
55:1169—1173.

Gould, S. J. 1977. Ontogeny and Phylogeny. Belknap
Press, Cambridge.

Gretillat, S. 1962. Etude du cycle évolutif du schistosome
des Ruminants domestiques de l'uest Africain et con-

firmation de 1'Espéce Schistosoma curassoni Brumpt
1931. Ann. Parasit. Hum. Comp. 37:557—568.

Gretillat, S. 1964. Differences morphologiques entre
Schistosma bovis (Souche de Karthoum) et Schistosoma
curassoni (Souche de Mauritanie). Rev. Elev. Med.

vét. Pays Trop. 17:429—432.

Haley, A. J. 1962. Role of host relationships in the
systematics of helminth parasites. J. Parasitol.
48:671—678.

Hecht, M. K. and .I. In Edwards. 1977. The nethodology
of phylogenetic inference above the species level.
Pp. 3-51 1g, Major Patterns of Vertebrate Evolution
(M. K. Hecht, P. C. Goody, and B. M. Hecht, eds.).
Plenum Press, New York.

Hennig, W. 1966. Phylogenetic Systematics. Univ. of
Illinois Press, Urbana.

Heyneman, D. 1960. On the origin of complex life cycles
of digentic trematodes. Pp. 133-152 In Libro Homenaje
a1 Dr. Edwardo Caballero y Caballaro. Sec. Educ.
Pub1., Int. Politic. Nac., Esc. Nac. Cienc. Biol.,
Mexico, D.F.

Holmes, J. C. 1983. Evolutionary relationships between
parasitic helminths and their hosts. Pp. 161-185
In Coevolution (D. J. Futuyma and M. Slatkin, eds.).
Sinauer Associates, Sunderland.

Hsu, H. F. and S. Y. Li Hsu. 1957. On the intraspecific
and interstrain variations of the male sexaul glands
of Schistosoma japonicum. J. Parasitol. 43:456-463.

 

 

 

 

237

Hubendick, B. 1978. Systematics and comparative morphology
of the Basommatophora. Pp. 1-48 _lp The Pulmonates
(V. Fretter and J. Peake, eds.). Academic Press

London.

Humphries, C. J. 1981. Biogeographical methods and the
southern beeches (Fagaceae: Nothofagus). Pp. 177—207
,In Advances in Cladistics (V. A. Fund and D. R.
Brooks, eds.). N. Y. Botanical Garden, Bronx.

Hussey, K. 1941. Comparative embryological development
of the excretory system in digenetic trematodes.
Trans. Amer. Micros. Soc. 60:171-210.

Inglis, W. G. 1971. Speciation in parasitic nematodes.
Adv. Parasitol. 9:185—223.

Ito, Y. 1960. Studies on Trichobilharzia corvi (Yamaguti,
1941). I. Morphological studies on the adult. Jap.
J. Parasitol. 9:556-563.

Janicki, C. V. 1920. Grundlinien einer "Cercomer Theorie”
zur Morphologie der Trematoden und Cestoden. Festsch.
fur Zschokke, Basel.

Johnston, S. J. 1917. On the trematodes of Australian
birds. J. Proc. Roy. Soc. N. S. Wales 50:187—261.

Kennedy, C. R. 1975. Ecological Animal Parasitology.
John Wiley and Sons, New York.

Khalifa, R. 1972. Studies on Schistosomatidae Looss,
1899 (Trematoda) of aquatic birds of Poland. I.
On the life cycle of Bilharziella polonica Kowalewski
1895, with a discussion of the subfamily Bilharziel-

linae Price 1929. Acta Parasitol. Pol. 20:343-365.

Khahl, M. 1924. On the morphology of Schistosoma bovis.
J. Helminthol. 2:81-86.

Kinsella, J. M. 1971. Growth, development and intra-
specific variation of Quingueserialis guinguerseralis
(Trematoda: Notocotylidae) in rodent hosts. J.
Parasitol. 57:62-70.

Kohn, A. 1964. S6bre o genéro Macrobilharzia. Travassos
1922 (Trematoda, Schistosomatidae). Mem. Inst.

Osw. Cr. 62:1-6.

Kollar, E. J. and C. Fisher. 1980. Tooth induction in
chick epithelium: Expression of quiscent genes for
enamel synthesis. Science 207:993-995.

 

238

Komiya, Y. 1939. Die Entwicklung des Exkretionssystems
einiger Trematodenlarven aus Alster und Elbe, nebst
Bemerkungen fiber ihren Entwicklungszyklus. Z. .Para—

sitenkd. 10:340-385.

Kowalewski, M. 1886. Studya Helminibologiszne. III.
Bilharzia polonica sp. nov. Rozpr. Akad. Umiejet.
31:41-70.

Kruatrachue, M., M. Bhaibulaya and C. Harinasuta. 1965.
Orientobilharzia harinasutai sp. nov., a mammalian
blood flulke, its morphology and life cycle. Ann.
Trop. Med. Parasitol. 59:181-188.

Kuntz, R. E. 1950. Embryonic development of the excretory
system in fork-tailed cercariae of the schistosomes
and in a blunt-tailed brachylaemid cercaria. Trans.
Amer. Micros. Soc. 64:1-20.

Kuntz, R. E. 1955. Biology of the schistosome complexes.
Am. J. Trop. Med. Hyg. 4:383-412.

Lal, M. B. 1937. Studies on the trematode parasites
of birds. Part II. Morphology and systematic position
of some new blood flukes of the family Schistosoma-
tidae. Proc. Ind. Acad. Sci. 6:274—283.

LaRue, G. R. 1957. The classification of digenetic Trema-

toda: a review and a new system. Exp. Parasitol.
6:306344.

Lee, H. F. 1962. Life history of Heterobilharzia americana
Price 1929, a schistosome of the raccoon an other

mammals in southeastern United States. J. Parasitol.
48:728—739.

Leigh, w. H. 1955. The morphology of Gigantobilharzia
huttoni (Leigh, 1953) an avian schistosome with marine

dermatitis—producing larvae. J. Parasitol. 41:262-
269.
LeRoux, P. L. 1933. A preliminary note on Bilharzia

margrebowiei, a new parasite of ruminants and possibly
man in northern Rhodesia. J. Helminthol. 11:57—62.

Liu and Bai. 1976. On bird schistosomes from Jilin Pro-
vince Jilinobiharzia crecci gen. nov. sp. nov.

(Schistosomatidae: Bilhariellinae) with a discussion
of the taxonomy of the subfamily Bilharziellinae.
Acta Zoologia Sinica 22:385—392.

239

Llewellyn, J. 1965. Evolution of parasitic plalyhelminths.
Pp. 47-78 lg Evolution of Pa

rasites (A. E. R. Taylor
ed.). Blackwell Scientific Publications,

Loss, A. 1899. Weiters Beitrage zur Kenntnis der Trema-
toden—Fauna Aegyptens. Zod. Jb. 12:521-784.

MacHattie, C. 1934. A prelimi

of Schistosoma turkestani
Roy.

nary note on the life history

cum Skryabin, 1913. Trans.
Soc. Trop. Med. Hyg. 30:115-124.

MacKenzie, D. I. and C. E. McKenzie. 1980. Morphological

variation in Plagiorchis noblei Park 1936 (Trematoda:

Plagiorchiidae) from Tyrannus tyrannus and T. verti-
calis (Aves: Tyrannidae). J. Parasitol. 66:145—149.

Macko, I. K. 1961. Revision of the species Ornithobil-
harzia cgnaliculata (Rudolphi, 1819) (Trematoda:
Schistosomatiae). Helminthologia 4:303—316.

Madhavi, R. and K. H. Rao. 1970. Orchispirium heterovitel-
latum gen. et. sp. n. (Trematoda: Sanguinicolidae)

from the ray fish, Dasyatis imbricatus Day, from
Bay of Bengal. J. Parasitol. 56:41-43.

Malek, E. A. 1977. Geographical distribution, hosts,
and biology of Schistosomatium douthitti (Cort, 1914)
Price, 1931. Can. J. Z001. 55:661-671.

Martin, W. E. 1944. Cercaria, solemyae n. sp., probably

a blood fluke, from the marine pelecypod, Solemya
velum. J. Parasitol. 30:191-195.

Martin, W. E. 1952. Another annelid first intermediate

host of a digenetic trematode. J. Parasitol. 38:356—
359.

McCoy, O. R. 1929. Notes on Cercariae
J. Parasitol. 15:199—208.

from Missouri.
McFarland, W. N., F. H. Plough, T. J. Cade and J. B. Heiser.

1979. Vertebrate Life. Macmillan Publishing Company,
New York.

McKenna, M. C. 1975. Toward a phylogenetic classification
of the Mammalia. Pp. 21-46 lg Phylogeny of the

Primates (W. P. Luckett and F. S. Szalay, eds). Plenum
Press, New York.
McLeod, J. A. 1937. Two new schistosomid trematodes

from water birds. J. Parasitol. 23:456-466.

 

 

‘3‘? _...—._. ______i.

240

McLeod, J. A. 1940. Studies on cercarial dermatitis

and the trematode family Schistosomatidae in Manitoba.
Can. J. Res. 18:1-28.

McMullen, D. B. and P. C. Beaver. 1945. Studies on schi-
stosome dermatitis. IX. The life cycle of three
dermatitisproducing schistosomes from birds and a
discussion of the subfamily Bilharziellinae (Trematoda:
Schistosomatidae). Amer. J. Hyg. 43:128-154.

Mehra, H. R. 1939. New blood flukes of the family Spiror—
chidae Stunkard (Trematoda) from the marine turtle
Chelone mydas of the Arabian sea with observations
on the synonymity of certain genera and classification
of the family. Proc. Nat. Acad. Sci. India 9:155-167.

Mehra, H. R. 1940. A new distome Enterohaematotrema
n. g. and a new blood fluke Hemiorchis bengalensis
n. sp. belonging to the family Spirorchidae (Stunkard)
and a new species of the genus Dendritobiharzia Skrja-
bin and Zakharrow, belonging to the family Schisto—
somatidae Poche, with remarks on the evolution of
the blood flukes. Proc. Nat. Acad. Sci. India 10:100—

118.

Mehra, H. R. 1950. Evolution of the blood flukes and
strigeid trematodes. Proc. Nat. Acad. Sci. India
19:9-38.

Mickevich, M. F. 1981. Quantitative phylogenetic bio-
geography. Pp. 209—222 In Advances in Cladistics,
Volume 1 (V. A. Funk and .D. R. Brooks, eds.). N.
Y. Botanical Garden, Bronx.

Mitter, C. and D. R. Brooks. 1983. Phylogenetic aspects
of coevolution. Pp. 65-98 lg Coevolution (D. J.
Futuyma and M. Slatkin, eds). Sinauer Associates,
Sunderland.

Miyairi, K. and Suzuki, M. 1913. On the development
of Schistosoma japonicum. Trop. Dis. Bull. 3:289-290.

Montgomery, R. E. 1906. Observations on bilharziosis
among animals in India. II. J. Trop. Vet. Sci.
1:138—174.

Monquin-Tandon, A. 1860. Elements de zoologie médicale.
Paris.
Mudaliar, S. VK and (L Ramanujachary. 1945. Schistosoma

nairi n. sp. from an elephant. Ind. Vet. J. 22:1—4.

241

Najim, A. TE 1950. Gigantobilharzia huronensis sp. nov.,
bird blood—fluke from goldfinch. J. Parasitol. 36:19.

Najim, A. T. 1956. Life history of Gigantobilharzia

huronensis Najim, 1950. A dermatitis producing bird
blood fluke (Trematoda: Schistosomatidae). Parasitol.
46:443-469.

Nelson, G. 1978. Ontogeny, phylogeny, paleobiology and
the biogenetic law. Syst. 2001. 27:324-345.

Nelson, G. and N. I. Platnick. 1981. Systematics and
Biogeography. Columbia Univ. Press, New York.

Noble, E. It and G. A. Noble. 1976. Parasitology. Lea
and Febiger, Philadelphia.

Odhner, T. 1910. Gigantobilharzia acotylea n. g. n.
sp. ein mit den Bilharzien vervandter Blutparasit

von enormer Lange. Zool. Anz. 35:380-385.

Odhner, T. 1912. Zum naturlichen System der digenen
Trematoden. 5. Die Phylogenie des Bilharzia-Typus.
Zool. Anz. 41:54-71.

Oglesby, L. C. 1961. A new cercaria from an annelid.
J. Parasitol. 47:233-236.

Olivier, L. 1940. Life history studies on two strigeid
tematodes of the Douglas Lake region, Michigan. J.
Parasitol. 26:447-477.

Olsen, 0. W. 1974. Animal Parasites. University Park
Press, Baltimore.

Platnick, N. I. and V. A. Funk, eds. 1983. Advances
in Cladistics, Volume 2. Columbia Univ. Press, New
York.

Poche, F. 1907. Einige Bemerkungen zur Nomenclatur den
Trematoden. Zool. Anz. 31:124-126.

Pojmanska, T. 1967. Variability of Leucochloridium para-
doxum Carus (ég. heckerti Kagan, 1952) (Trematoda:
Brachylaimidae) in natural and experimental
conditions. Acta Parasitol. Pol. 14:381-398.

Price, E. W. 1929. A synopsis of the Trematode family
Schistosomidae with descriptions of new genera and
species. Proc. U.S. Nat. Mus. 75:1-39.

 

242

Price, E. W. 1931. Note on Macrobilharzia Travasos.
J. Parasitol. 17:230-231.

Price, E. W. 1934. New genera and species of blood flukes
from a marine turtle, with a key to the genera of
the family Spirorchidae. J. Wash. Acad. Sci. 24:132-
141.

Price, E. W. 1943. A redescription of Heterobilharzia
americana Price (Trematoda: Schistosomatidae). Proc.
Helminthol. Soc. Wash. 10:85—86.

Price, H. F. 1931. Life history of Schistosomatium dou—
thitti (Cort). Amer. J. Hyg. 13 685-727.

 

Price, P. W. 1980. Evolutionary biology of parasites.
Princeton Univ. Press, Princeton.

Purchon, R. D. 1977. The biology of the mollusca. Perga-
mon Press, Oxford.

Rensch, B. 1959. Evolution above the species level.
Columbia Univ. Press, New York.

Rios Leite, A. C., H. M. de Araiyo Costa and J. 0. Costa.
1978. Trichobilharzia jeguitibaensis np. n. (Trema-

toda, Schistosomatidae) in Cairina moschata domestica

(Anatidae). Rev. Brasil Biol. 38:843—846.

Rosen, D. 1978. Vicariant patterns and historical explana-
tion in biogeography. Syst. Zool. 27:159-188.

Ross, G. C., V. R. Southgate and R. J. Knowles. 1978.

Observations on some isoenzymes of strains of Schisto-
soma bovis, S. matheei, S. margrebowiei, and S. lei-

peri. z. Pargsitenkd. 57:49—56.

Rudolphi, C. A. 1819. Entozoorum synopsis cui accedunt
mantissa duplex et indices locupletissimi. Berolini.

Schell, S. C. 1970. How to know the trematodes. W.
M. C. Brown Co., Dubuque.

Schiller, E. L. 1959a. Experimental studies (n1 morpho-
logical variation in the cestode genus Hymenolepis.
I. Morphology and development of the Cysticercoid
of Hymenolepis nana in Iiipgligm ggnfgggm. Exp.
Parasitol. 8:91-118.

Schiller, E. L. 1959b. Experimental studies on morpho—
logical variation in the cestode genus Hymenolepis.
II. Morphology and development of the strobilate

243

phase of Hymenolepis nana in different mammalian
host species. Exp. Parasitol. 82215-235.

Schiller, E. L. 1959c. Experimental studies on. morpho-
logical variation in the cestode genus Hymenolepis.
III. X-irradiation as a mechanism for facilitating
analyses in S. nana. Exp. Parasitol. 8:427-470.

Schwetz, J. 1951. A comparative morphological and bio-
logical study of Schistosoma haematobuim, .§. bovis,
S. intercalatum Fisher, 1934; S. mansoni and S. rod-
haini Brumpt, 1931. Ann. Trop. Med. Parasitol. 45:92—
98.

 

Short, R. B. 1983. Presidential address: Sex and the
single schistosome. J. Parasitol. 69:3-22.

Short, R. B. and R. B. Holliman. 1961. Austrobilharzia

penneri, a new schistosome from marine snails. J.
Parasitol. 47:447—452.

Simpson, G. G. 1945. The principles of classification
and a classification of mammals. Bull. Amer. Mus.
Nat. Hist. 85:1—350.

Sinha, P. K. and H. D. Srivastava. 1956. Studies on
Schistosoma incognitum Chandler, 1926. I. On the
synonomy and morphology (If the blood fluke. Parasi-
tology 46291-100.

Sinha, P. K. and H. D. Srivastava. 1960. Studies on
Schistosoma incognitum Chandler, 1926. II. On the
life history of the blood fluke. J. Parasitol. 46:629-

641.

Skrjabin, K. I. 1913. Schistosomum turkestanicum nov.
sp., ein neuer parasit des Rindes aus Russisch-Turke-
stan. Zeit. Inf. Kran. Parasit. Krank. Hyg. Haus.
13:457—469.

Skrjabin, K. 1., ed. 1951. Trematodes of animals and

man. Principles of trematodology, Vol. V. Izdatelstro
Akademii Nauk SSSR, Moscow.

Smith, .L W. 1972. The blood flukes (Digenea: Sanguini-
colidae and Spirorchidae) of cold-blooded vertebrates

and some comparisons with the schistosomes. Hel-
minthol. Abstr. A412161-204.

Sneath, P. H. A. and R. R. Sokal. 1973. Numerical taxonomy
W. H. Freeman and Co., San Francisco.

 

244

Stiles, C. W. and A. Hassall. 1898. An inventory of
the genera and subgenera of the trematode family
Fasciolidae. Arch. Parasitol. 1:81—99.

Straney, D. O. 1980. Relationships of phyllostomatine
bats: Evaluation of phylogenetic hypotheses. Ph.D.
thesis, Univ. California, Berkeley.

Stunkard, H. W. 1921. Notes on North American blood
flukes. Amer. Mus. Novit. 12:1—5.

Stunkard, H. W. 1923. Studies on North American blood
flukes. Bull. Amer. Mus. Nat. Hist. 75:165-221.

Stunkard, H. W. 1957. Host—specificity and parallel
evolution of parasitic flatworms. Z. Tropenmed.
Parasitol.

Stunkard, H. W. 1970. Trematode parasites of insular

and relict vertebrates. J. Parasitol. 56:1041-1054.

Stunkard, H. W. and M. C. Hinchcliffe. 1952. The morpho-
logy and life history of Microbilharzia variglandis
(Miller and Northrup, 1926). Stunkard and Hinchcliffe,
1951, avian blood—flukes whose larvae cause "swimmers
itch” of ocean beaches. J. Parasitol. 38:242-265.

Szidat, L. 1929a. Die Parasiten des Hausgeflugels. 3.
Bilharziella polonika ng., ein im Blut schmarozender
Trematode unserer Enten, seine Entwicklung und Ueber-
tragung. Archiv. Gefluegelkunde 3:78-87.

Szidat, L. 1929b. Zur Entwicklungsgeschechte des Blut-
terematoden der Enten, Bilharziella polonica Kow.
Zbl. Bakt. I Orig. 3:461-470.

Tada, S. 1928. On the development and structure of Schi-
stosoma japonicum in the final host. Okayama Igakkai
Zasshi 463:1-36.

 

Tanabe, B. 1923. The life-history of a new schistosome,
Schistosomatium pathlocopticum Tanabe, found in experi-
mentally infected mice. J. Parasitol. 9:183—198.

Taylor, A. E. R., ed. 1965. Evolution of Parasites.
Blackwell Sci. Pub., Oxford.

Taylor, A. E. R. and R. Muller, eds. 1976. Genetic aspects
of host-parasite relationships. Blackwell Sci. Pub.,
Oxford.

 

244

Stiles, C. W. and A. Hassall. 1898. An inventory of
the genera and subgenera of the trematode family
Fasciolidae. Arch. Parasitol. 1:81—99.

Straney, D. O. 1980. Relationships of phyllostomatine
bats: Evaluation of phylogenetic hypotheses. Ph.D.
thesis, Univ. California, Berkeley.

Stunkard, H. W. 1921. Notes on North American blood
flukes. Amer. Mus. Novit. 12:1-5.

Stunkard, H. W. 1923. Studies on North American blood
flukes. Bull. Amer. Mus. Nat. Hist. 75:165-221.

Stunkard, H. W. 1957. Host—specificity and parallel
evolution of parasitic flatworms. Z. Tropenmed.
Parasitol.

Stunkard, H. W. 1970. Trematode parasites of insular

and relict vertebrates. J. Parasitol. 56:1041—1054.

Stunkard, H. W. and M. C. Hinchcliffe. 1952. The morpho-
logy and life history of Microbilharzia variglandis
(Miller and Northrup, 1926). Stunkard and Hinchcliffe,
1951, avian blood-flukes whose larvae cause "swimmers
itch" of ocean beaches. J. Parasitol. 38:242-265.

Szidat, L. 1929a. Die Parasiten des Hausgeflugels. 3.
Bilharziella polonika £21., ein im Blut schmarozender
Trematode unserer Enten, seine Entwicklung und Ueber-
tragung. Archiv. Gefluegelkunde 3:78—87.

Szidat, L. 1929b. Zur Entwicklungsgeschechte des Blut-
terematoden der Enten, Bilharziella polonica Kow.
Zbl. Bakt. I Orig. 3:461-470.

Tada, S. 1928. On the development and structure of Schi-
stosoma japonicum in the final host. Okayama Igakkai
Zasshi 463:1-36.

Tanabe, B. 1923. The life-history of a new schistosome,
Schistosomatium pathlocopticum Tanabe, found in experi—
mentally infected mice. J. Parasitol. 9:183-198.

Taylor, A. E. R., ed. 1965. Evolution of Parasites.
Blackwell Sci. Pub., Oxford.

Taylor, A. E. R. and R. Muller, eds. 1976. Genetic aspects
of host-parasite relationships. Blackwell Sci. Pub.,
Oxford.

 

 

245

Thurston, J. P. 1963. Schistosomes from Hippopotamus

amphibius L. I. The morphology of Schistosoma hippopo-
tami sp. nov. Parasitology 53:49-54.

Thurston, J. P. 1964. Schistosomes from Hippopotamus
amphibius L. 11. The morphology of Schistosoma edward-
iense sp. nov. Parasitology 54:67-72.

Travassos, L. 1922. Informacées sdbre a fauna helmin—
tologica de Mato Grosso. Folha Med. 3:187-190.

Travassos, L. 1942. Contribuicées ao conhecemento dos
"Schistosomatidae." Sobre "Ornithobilharzia canali-
culata" (Rudolphi, 1819). Rev. Brasil Biol. 2:473—
476.

Tubangui, M. A. 1933. Trematode parasites of Philippine
Vertebrates, VI. Descriptions of new species and
classification. Philipp. J. Sci. 52:167-197.

Ulmer, M. J. and F. J. Vande Vusse. 1970. Morphology
of Dendritobilharzia pulverulenta (Braun, 1901) Skrja-
bin, 1924 (Trematoda: Schistosomatidae) with notes
on secondary hermaphroditism in males. J. Parasitol.
56:67—74.

Van der Land, J. 1967. A new blood fluke (Trematoda)
from Chimaera monstrosa L. Proc. Koninkl. Neder.
Akad. van Wetenschappen Amsterdam 70:110-120.

Vande Vusse, F. J. 1980. A review of the genus Dendrito-
bilharzia Skrjabin and Zakhorow 1920. (Trematoda:
Schistosomatidae). J. Parasitol 66:814—822.

Van Tyne, J. and A. J. Berger. 1976. Fundamentals of
Ornithology. John Wiley and Sons, New York.

Vogel, V. H. and W. Minning. 1940. Bilharziose bei Ele-
phanten. Arch. Schiffs—u. Tropenhyg. 44:562-574.

Wakelin, D. 1976. Host Responses. Pp. 115-141 lp Ecolog-
ical Aspects of Parasitology (C. R. Kennedy, ed.).
North—Holland Pub. Co., Amsterdam.

Wales, J. H. 1958. Two new blood fluke parasites of
trout. Calif. Fish Game 44:125—136.

Ward, H. B. 1921. A new blood fluke from turtles. J.
Parasitol. 7:114-128.

Warren, K. S. 1973. Schistosomiasis: The Evolution of
a Medical Literature. Mass. Inst. Tech., Cambridge.

 

 

 

246

Watertor, J. L. 1967. Intraspecific variation of adult
Telorchis bonnerensis (Trematoda: Telorchiidae) in
amphibian and reptilian hosts. J. Parasitol. 53:962-—
968.

Weinland, D. F. 1858. Human Cestoides. Cambridge.

Wetzel, R. 1930. A new species of trematode worm of
the genus Ornithobilharzia from a Canada goose. Proc.
U.S. Nat. Mus. 78:1-4.

Wiley, E. O. 1981. Phylogenetics: the theory and practice

of phylogenetic systematics. John Wiley and Sons,
New York.

Witenberg, G. and J. Lengy. 1967. Redescription of Orni-
thobilharzia canaliculata (Rud.) Odhner, with notes
on classification of the genus Ornithobilharzia and

the subfamily Schistosomatinae (Trematoda). Israel
J. 2001. 16:193-204.

Wright, C. A. 1960. Relationships between trematodes
and molluscs. Ann. Trop. Med. Parasitol. 54:1-7.

Yamaguti, S. 1971. Synopsis of Digenetic Trematodes
of Vertebrates, Vol. I and II. Keigaku Pub. Co.,
Toyko.

Young, R. T. 1937. Another record of avian schistosomes
in North America. J. Parasitol. 23:295-296.

Zilch, A. 1959. Gastropoda, Vol. 2, Euthyneura. Pp.
1-834 1p Handbook Palaozool. (O. H. Schindewolf,
ed.).

 

 

 

 

 

 

 

 

 

      

‘ MIC

11171111111111![ilirflflflflilfliflfilflfil“