The origins of parasitism in the platyhelminthes

Internafional

Pergamon

Journal

for Pamirology, Vol. 24. No. 8, pp. 1099-l 115, 1994 Copyright 0 1594 Australian Society for Parasitology Else&r Science Ltd Printed in Great Britain. All rights reserved lxJxb75 19/94 $7.00 + 0.00

THE ORIGINS OF PARASITISM IN THE PLATYHELMINTHES K. ROHDE Department of Zoology, University of New England, Armidale, NSW 2351, Australia Abstract-Rohde K. 1994. The origins of parasitism in the Platyhelminthes. International Journal for Parasitology 24: 1099-1115. Symbiotic associations have arisen independently in several groups of the largely free-living turbellarians. Morphological adaptations of turbellarians to a symbiotic way of life include suckers and adhesive glands for attachment, elaborate systems of microvilli and other epidermal structures for absorption of food, glands for the formation of cysts, cocoons and cement material, and lack of a pharynx and intestine in some species. However, many species closely resemble their free-living relatives. Egg production is greatly increased at least in some species, and life cycles are always direct. Food of symbiotic turbellarians consists of host food and/or host tissue. Ectosymbiotes show fewer physiological adaptations than entosymbiotes. The major groups of parasitic Platyhelminthes (Trematoda Aspidogastrea, Trematoda Digenea, Monogenea, Udonellidea, Cestoda including Gyrocotylidea, Amphilinidea and Eucestoda), form one monophylum, the Neodermata, characterized by a neodermis (tegument) replacing the larval epidermis, epidermal cilia with a single horizontal rootlet, sensory receptors with electron-dense collars, spermatozoa with axonemes incorporated in the sperm body by proximodistal fusion, and protonephridial flame bulbs formed by two cells each contributing a row of longitudinal ribs to the filtration apparatus. The sister group of the Neodermata is unknown but is likely to be a large taxon including the Proseriata and some other turbellarian groups. Among the Neodermata, the Aspidogastrea is likely to be the most archaic group, as indicated by DNA studies, morphology, life cycles and physiology. Aspidogastreans can survive for many days or even weeks outside a host in simple media, they show little host specificity, and have an astonishin~y complex nervous system and many types of sensory receptors, both in the larva and the adult. It is suggested that Aspidogastrea were originally parasites of molluscs (and possibly arthropods and other invertebrates) and that they are archaic forms which have remained at a stage where vertebrates represent facultative hosts or obligatory final hosts into which only the very last stages of the life cycle (maturation of the gonads) have been transferred. The complex life cycles of Digenea have evolved from the simple aspidogastrean ones by intercalation of multiplicative larval stages (sporocysts, rediae) in the mollusc host, and of cercarial stages ensuring dispersal to the now obligatory final host. Monogenea may have lost the molluscan host or evolved before the early neodermatans had acquired it. Cestoda either replaced the original molluscan with an arthropod host, retained an original arthropod host or evolved from an early neodermatan before molluscan hosts had been acquired, newly acquiring an arthropod host. Horizontal gene transfer and implications for mosaic evolution in the Platyhelminthes are discussed. INDEX KEY WORDS: Platyhelminthes; parasitism, evolution; phylogeny; Aspidogastrea; Turbellaria; Fecampiidae; horizontal gene transfer; DNA.

INTRODUCTION

The Platyhelminthes is a large phylum, including probably about 10,000 described species. It is likely that the majority of species is still unknown. Thus, few of the meiofaunal turbellarians of Australian beaches have been described, and only a small proportion of the very rich fauna of monogeneans of Australian marine fishes is known (Rohde, 1976). Almost certainly, the same applies to free-living and parasitic flatworms of many countries, particularly those in the tropics and subtropics. Platyhel~nthes are conventionally divided into the largely free-living turbellarians and a number of

parasitic classes, that is, the trematodes, monogeneans, and cestodes (the latter including the amphilinids, gyrocotylids, and eucestodes). The taxonomic position of the udonellids is not clear, but it is likely that they stand close to the monogeneans and/or cestodes (Rohde & Watson, 1993b). Ehlers (1984, 1985), using many light-microscopic and ultrastructural data, has shown that the major parasitic groups form one monophylum, the Neodermata. Except for some taxa which are secondarily modified, they all share a neodermis (= tegument) with subtegumental perikarya connected to the syncytial surface layer by several branching

1099

1100

K. ROHDE

Fig. 1. Chara~terjst~cs of the Neode~ata. (A) Neodermis (= te~ent, t) of larval A~tru~~hi~~a ei~~guzu {Amphil~idea) below larval epidermis (ep) and connected to subtegnmental perikaryon (p) by branching cytoplasmic processes (arrowheads). (B) Single horizontal rootlet (r) of epidermal cilium of larval A. efo~uta. ep = epidermis. (C) Oblique cross-section through flame bulb of larval ~~~ticotyZe purvisi (Aspidogastrea). Note the hltration apparatus with a row of external (ex) and internal ribs (in) connected by a “membrane” of extracellular matrix (arrowheads); cilia (c) of flame tightly packed. (D) Sensory receptors of Lcbutostoma manteri (Aspidogastrea). Note electron-dense collars (e), rootlet of cilinm (r) and tegument @). Scale bars 1 pm (A), 0.5 pm (B-Q).

Origins of parasitism in Platyheltninthes processes, which replaces the larval epidermis (Fig. IA). They also share epidermal cilia with a single, horizontal rootlet (Fig. IB), sensory receptors that have electron-dense collars (Fig. lD), spermatozoa that originally have 2 axonemes incorporated in the sperm body by proximo-distal fusion (Fig. 2A, B) (Justine, 1991), and protonephridial flame bulbs formed by 2 cells, the terminal and proximal canal cells, each contributing a row of longitudinal ribs (rods) to the filtration apparatus (weir) of the flame bulb (Fig. 1C). All the extant Neodermata are parasitic, either living in or on a host. SYMBIOSIS IN THE ~TURB~L~RIA” According to Ehlers (1985), the tur~llarians are a paraphyletic group, and strictly taken, using phylogenetic systematics (cladistics), the term “Turbellaria” should not be used. I use it, in the following, as a name for an agglomerate of largely free-living platyhelminth taxa (all the non-neodermatans) without implying monophyly of these taxa. Cannon’s (1986) guide to families and genera of Turbellaria shows that symbiotic relationships among the turbellarians are very common. In most cases, our knowledge is insufficient for a decision on whether the relationship is commensal or parasitic. It therefore seems appropriate to use the more general term symbiotic (sensz4l&o, including all cases of close associations between species). Only among the Catenulida, Macrostomida, Haplopharyngida, and Lecithoepitheliata are there no known species that live in symbiosis with other, larger species, although cases of symbiosis with smaller protozoans and other microorganisms are known. Thus, Ott, Rieger, Rieger & Enders (1982) found bacteria in a catenulid, we found protozoans living in the lecithoepitheliate Prorhynchus (in unpublished observations), and it is likely that detailed eleetronmicroscopic studies of other turbellarians will reveal many further cases. For examples of symbiosis of turbella~ans with smaller species see Taylor (1971), Jennings (19X9), Noury-Srafri, Justine & Euzet (1989), Saffo (1992), and Anderson, Newman & Lester (1993). Most turbellarian groups contain at least some species that are associated with larger species.

Fig. 2. Characteristics of the Neodermata. A. Crosssections through sperm of Lobatostoma manteri (Aspidogastrea). Note two axonemes (ax) incorporated in sperm body. nu - nucleus, m - mitochondrion. B. Longitudinal section through developing spermatozoon. Two flagella (0, each with a rootlet (r) and separated by an inter~nt~olar body (i) are still free. They later fuse with the median cytoplasmic process (me) by proximodistal fusion (beginning at the base of the flagella). Scale bars 0.5 pm.

1101

1102

K. ROHDE

Among the Nemertodermatida, Meara lives in holothurians, but is probably an endocommensal feeding on gut contents. The acoelan Aechmalotus also lives in holothurians and the prolecithophorans ~r~storn~ (possibly not a prol~ithophoran, e.g. Fleming, 1986) and ~chthyophug~ are symbionts of molluscs and fish, respectively. Among the Proseriata, some Monocelididae are symbiotic, the rhabdocoel Genostoma lives on small crustaceans (Hyra, 1993), and among the Dalyelliida, Acholades lives in the tube feet of sea stars, and some species of the ~alyelliida, for example Djdymorchis (possibly a temnocephalan), are symbiotic in the gill chamber of freshwater crayfish. The dalyelliids Graffilla and Paravortex parasitize molluscs, Hypoblepharina lives in symbiosis with crustaceans, and some Provorticidae, all Pterastericolidae (e.g. Jondelius, 1989), all Umagillidae (e.g. Cannon, 1982, 1987) and the Temnocephalida (e.g. Jones & Lester, 1992) are symbiotic. Typhloplanida, Kalyptorhynchia, Tricladida (some Bdellouridae), and Polycladida (Emprosthopharynx associated with hermit crabs, Ap~dioplanu with gorgonians), also include some symbiotic species. All Fecampiidae are endoparasitic. Lists and brief discussions of turbellarian families containing symbiotic species were given by Jennings (1974b) and Burt (1988). Jennings (1974b) refers to 120 species of 27 families that live symbiotically, most with echinoderms and crustaceans. At the gross mo~hological level, some of the symbiotic species show distinct adaptations to their way of life. The Fecampiidae and Acholadidae, for example, lack a pharynx and intestine, and the Temnocephalida have posterior suckers for attachment to a host (Joffe, 1988). Electron-microscopic studies have demonstrated the presence of adhesive glands in the Pterastericolidae (see Jondelius, 1992), and an elaborate system of microvilli and other epidermal structures facilitating absorption of food in the Fecampiidae (Bresciani & K&e, 1970; Williams, 1990a,b, 1991), Acholadidae {Jennings, 1989) and Umagillidae (Jondelius, 1986). Also in the Fecampiidae, glands for the formation of a cyst on the cuticle of the crustacean host have been demonstrated (Ksie & Bresciani, 1973), as well as subepidermal glands probably secreting the cocoon and cement material (Blair & Williams, 1987; Williams, 1990b). The loss of a vertical ciliary rootlet of epidermal cilia in some umagillids may be an adaptation to a parasitic way of life, since it is also found in all Neodermata (see below). The species Syndisyrinx punicea from the intestine of the sea urchin Heliocidaris erythrogramma has “normal” vertical and horizontal rootlets, Cle~stog~rn~ longi-

cirrus from the intestine of the sea cucumber Stichopus variegatus has a single horizontal rootlet

with a thin but long, more vertically directed branch (possibly a rudimentary vertical rootlet), and Seritia stichopi from the intestine of Stichopus ch~oronot~ has a single horizontal rootlet without a side branch (Rohde, Watson & Cannon, 1988; Rohde & Watson, 1988). However, the functional significance of the loss is not known and it may well be that lack of a vertical rootlet in the Neodermata is an evolutionary accident, characteristic of the common ancestral species, but without any adaptive value. This does not seem unlikely in view of the fact that epidermal cilia of Neodermata are found only in (usually) freeliving larval stages. Other studies have provided no evidence that convergent evolution in turbellarians in the same, symbiotic habitat has led to similar mo~hologi~al characteristics. Thus, Rohde & Watson (1990a,b) compared various ultrastructural characteristics (epidermis, glands) of Didymorchis with other platyhelminths living in the gill chamber of freshwater crayfish, as well as with various free-living turbellarians, and concluded that there are no ~mmon characteristics identifiable at the electron-microscopic level, which enable them to live in their habitat. In particular, an epithelium, with the perikarya located below the surface epidermis, is not only found in Didymorchis, but in many free-living turbellarians, and it is not found in Temnocepha~a from the gill cavity of the same host. Nom-y-Srai’ri et al. (1989) found no obvious adaptations to parasitism in the ciliated epidermis of the rhabdocoel Paravortex living in the intestine of molluscs, and Holt ,& Mettrick (1975) noted the close morphological similarity of the ~agillid Syndesmis ~unciscuna with free-living turbellarians. With regard to egg production and development, some symbiotic turbellarians have been shown to produce very large numbers of eggs, much greater than free-living turbellarians and necessary to compensate for the loss of most offspring that do not succeed in infecting a host. Thus, Kanneworf & Christensen (1966) reported the deposition of more than 500,000 capsules, each containing two eggs, by one female Kronborgia caridicola (Fecampiidae). Life cycles may involve symbiotic and free-living stages, but never an intermediate host. For example, the umagillid dnoplodi~m hymanae parasitizes the body cavity of holothurians. Egg capsules are released into the coelom of the host where they become ensheathed by host coelomocytes forming masses of usually about 1 mm in diameter and containing up to several 100 egg capsules. The capsules leave the

Origins of parasitism in Platyhe~inth~s

host probably through small ducts connecting the coelom with the rectum; embryos can survive for 10-l 1 months in capsules. Hatching occurs when egg capsules are ingested by a host, induced by the host digestive fluid. Larvae penetrate the wall of the lower intestine or of the respiratory tree and mature in the coelom (Shinn, 1985a,b). Due largely to the studies of Jennings and collaborators (Jennings & Mettrick, 1968; Mettrick & Jennings, 1969; Jennings, 1971, 1974a,b, 1977, 1980, 1981,1988,1989; Jennings & LeFlore, 1979; Cannon & Jennings, 1987; Jennings & Cannon, 1987; Jennings & Hick, 1990) on nutrition and respiration, we have learned much on physiological adaptations of symbiotic turbellarians. For example, among three graffillid rhabdocoels, all feeding on their “hosts partially digested food plus the cellular debris released at the end of the hosts’ own digestive cycle”, one supplements its diet by ingesting host digestive cells and utilizes host enzymes for its own digestion (Jennings, 1981). In some cases, even related species show markedly different degrees of dependence on the host. Thus, among the umagillids, Syndisyrinx ~anci~cunus ingests host in~stinal tissue and symbiotic ciliates, Syndesmis dendrustrorum feeds on intestinal tissue and food ingested by the host, and Syndesmis echinorum lives entirely on host intestinal tissue (Shinn, 1981). Overall, Jennings (1988) concluded that “ectosymbiotes show fewer adap~tions than entosymbiotes and tend to resemble free-living turbellarians in most respects”; and generally, they use fragments of host food. Entosymbiotes increasingly depend metabolically on their hosts, using their food, digestive enzymes and oxygen. In spite of the adaptations to a symbiotic way of life, all symbiotic turbellarians either show clear relationships with free-living groups (most taxa) or stand completely on their own, without obvious relationships to other groups (Fecampiidae). The evidence clearly indicates that a symbiotic way of life has arisen repeatedly among several unrelated turbellarian groups. NEODERMATA

The vast majority of parasitic flatworms belongs to the trematodes (digeneans and aspidogastreans), monogeneans and cestodes (amphilinids, gyrocotylids and eucestodes). Monophyly of these major parasitic groups of flatworms, the Neodermata, originally proposed on the basis of morphological (mainly ultrastructural) evidence (Ehlers, 1984, 1985), has been verified by more recent work using techniques of molecular biology (RNA/DNA)

1103

sequencing (Baverstock, Fielke, Johnson, Bray & Beveridge, 1991; Blair, 1993; Rohde, Hefford, Ellis, Baverstock, Johnson, Watson & Dittmann, 1993). THE SISTER GROUP (ADELPHOTAXON) OF THJZ~ODER~TA

Although there can be no doubt that all major groups of parasitic Platyhelminthes form one monophylum, the question of the sister group of this monophylum has not been settled. Most authors assume that it is to be found among the rhabdocoel tur~llarians, in particular the Dalyellioida (Ehlers, 1985), or the Temnocephalida (Brooks, 1989). According to Ehlers (1985), the synapomorphy shared by the supposed sister group and the Neodermata (both together comprising the Dolioph~~ophora) is a special type of pharynx, the pharynx doliiformis. However, Rohde (199Oa) has pointed out that few electron-microscopic studies of the pharynges of Neodermata and turbellarians have been made and that evidence for the homology of the pharynx of the “Doliopharyngiophora” is practically non-existent. Furthermore, Joffe (1987), on the basis of detailed light-microscopic studies of the pharynges of various parasitic and free-living flatworms (Joffe, Slusarev & Timofeeva, 1987; Joffe & Chubrik, 1988) concluded that the doliiform pharynx has evolved repeatedly in different groups, and that the parasitic taxa have not evolved from the pol~hyleti~ Dalyel~oida. Most importantly, all rhabdocoels have a characteristic type of flame bulb of the protonephiridial system [single row of longitudinal ribs, ribs with bundles of microtubules, lack of internal leptotriches, one perikaryon (at least usually) continuous with several flame bulbs, Rohde, 19911, which clearly is not plesiomorphic for the Neodermata. Hence, the rhabdocoels cannot be the “primitive” sister group of the Neodermata. Brooks (1989, earlier references therein) suggested that the Temnocephalida is the sister group of the Neodermata and, within the Neodermata, the Udonellidea is the sister group of all others (the Cercomeridea), and Jondelius & Thollesson (1993) proposed that the Pterastericolidae, Fecampiidae and Acholadidae share a last common ancestor with the Neodermata. Both hypotheses are based on superficial similarities, i.e. characters the homology of which has not been established, and are not supported by ultrast~ctural and DNA evidence (Rohde, 1990a; Rohde et al., 1993).

Ultrastructural evidence seems to indicate that the Fecampiidae, a small group of flatworms parasitic in marine crustaceans, may be the sister group of the

i 104

K. ROHDE

Neodermata. They are usually included in the rhabdocoels but lack of any s~apomorphies of Fecampiidae and Rhabdocoela indicates that this inclusion is without justification. Several characteristics show that they do not belong to the Neodermata: the ciliated epidermis of the free-swimming larva is retained by the parasitic adult (i.e. they lack a neodermis), the epidermal cilia at least of some species and at certain developmental stages have a vertical and horizontal rootlet, and the epidermal sensory receptors lack electron-dense collars. However, other characteristics may indicate a close relationship with the Neodermata: mature sperm iack dense bodies and have two fully incorporated axonemes Williams, 1988); in~o~oration of axonemes in the sperm body is proximo-distal (Watson & Rohde, 1993a,b); the eye is of the reflector type similar to that in polystome monogeneans (Watson, Williams & Rohde, 1992); and the flame bulb of the larval protonephridium has a rudimentary weir composed of two interdigitating cells and a septate junction along the canal cell, resembling those of the Monogenea and Trematoda (Watson, Rohde & Williams, 1992). Although these similarities are suggestive of a sister group relationship between Fecampiidae and Neodermata, convergent evolution cannot be ruled out, because inco~oration of axonemes in the sperm body is found in some other turbellarians (although always by distalproximal fusion), a reflector eye of a somewhat different type containing both pigment granules and crystalline platelets has been shown to occur in acoelans (Popova & Mamkaev, 1985; Yamasu, 1991), and reflector eyes are also known from some other invertebrates; flame bulbs resembling those of trematodes/monogeneans also occur in some turbellarians, e.g. the Proseriata (Rohde, 1991). A comparison of DNA sequences of the fecampiid ~ronbor~~~ isopo~i~o~~ and various free-living and parasitic Platyhelminthes suggests that the species is not a close relative of the Neodermata (Rohde, Luton, Baverstock &Johnson, 1994), and thus, the morphological similarities of the two groups appear indeed to be due to convergent evolution. Another candidate for the sister group of the Neodermata may be the Proseriata. Rohde (1988) tentatively suggested this on the basis of the very great similarity in the structure of their protonephridial flame bulbs. Recent DNA studies have shown that a large taxon comprised of several “turbellarian” groups including the Proseriata may be the sister group (Rohde et nl., 1993), i.e. that the major groups of parasitic plathelminthes have arisen early in the evolution of Platyhelminthes. However,

the DNA sequences used to date are relatively short (ca. 550 base pairs) and not all large platyhelminth taxa have been examined. Thus, I leave open the question of the exact composition of the sister group of the Neodermata. A tentative phylogenetic tree of the Platyhelminthes is given in Fig. 3. It shows that the phylogenetic relationships of most “turbellarian” taxa has not been resolved and that the sister group of the Neodermata cannot be clearly identified. ASPIDOGASTBEA, THE MOST “PRIMITIVE” MEODERMATA In contrast to the uncertainty about the closest relative of the Neodermata, there is strong support for the view that, among the Neodermata, the Aspidogastrea (Fig. 4) is the most archaic group. Some earlier authors have held the view that the Aspidogastrea is the most “primitive” group of trematodes (e.g. Stunkard 1917, 1967), and Rohde (1971) provided the followed evidence for this view: Aspidogastrea are poorly adapted to a parasitic way of life, as indicated by a very long survival time outside a host in simple media (water, saline solution), by a low host and organ specificity, an astonishingly complex nervous system and a great variety and number of sense receptors, and the small number of species interpreted as a measure of their relatively unsuccessful adaptation to their habitat. Rohde (1971) further concluded that: 1. Aspidogastrea are primitively parasites of mollusts, since for several species sexual maturation in molluscs has been shown; 2. Aspidogastrea are, of all parasitic Platyhelminthes, most closely related to the Digenea, since both groups have posterior suckers, posterior excretory pores, a Laurer’s canal and a similar ultrastructure of the ~otonephridial system; 3. Digenea are probably also primitive parasites of molluscs, a view supported by the stricter host specificity to the mollusc than the vertebrate host, and by their close phylogenetic relationship with the Aspidogastrea; 4. Aspidogastrea are archaic forms which have remained at a stage where the vertebrate represents a facultative host in which they may survive when ingested, whereas life in vertebrates has become necessary for Digenea: the sac-like caecum of most aspidogastreans and the low degree of oligomer~ation (reduction in the number of repetitive structures) are indicative of archaic forms, and for several species survival in vertebrates has been demonstrated;

Origins of parasitism in Platyhelminthes

1105

Fecampiida

-t Typhloplanoida

i Rhabdocoela

Dalyellioida (incl. Temnocephalida) .: Lecithoepaheliata H~~pha~ngida Pmi~hophom Polycladida Tricladida Macrostomida Proseriata Aspidogastrea Trematoda

i

Digenea Poiyopkthocotyiea Monogenea ~no~~~~lea i Neodermata

U~neliidea ? -I

Udoneilidea ?

i

Gyrocutylidea

; : Nephrexozeuktica 1 i

Amphilinidea Eucestcda

.i

.:

Fig. 3. Tentative phylogenetic tree of the Platyhelminthes, largely based on the ultrastructure of protonephridia and 18 S ribosomal DNA sequences. 5. The complex life cycle of the Digenea has evolved from the simple life cycle as found in Aspidogastrea by intercalation of multiplicative larval stages (sporocyts, rediae) in the mollusc host, and of cercarial dispersal stages. Recent studies lend support to the view that aspidogastreans are indeed archaic neodermatans. Blair’s (1993) DNA studies have led to the conclusion that the Aspidogastrea is “close to the base of either representing the sister the Neodermata”, group of the other trematodes, or of all other neode~atans. The first alternative is more likely in view of the finding by Rohde er al. (1993), who used partial sequences of 18 S ribosomal DNA (approximately 580 base pairs) of one nemertean, 20 free-

living and parasitic Platyhelminthes, and Homo and as outgroups, that aspidogastreans are the sister group of the other trematodes rather than of all other neodermatans. Similar morphological characteristics (posterior excretory pores, posterior suckers, Laurer’s canal), and similar hosts (molluscs, vertebrates) also suggest that aspidogastreans and digeneans are close relatives. Artemia

BIOLOGICAL AND MORPHOLOGICAL FEATURES OF THE ASPIDOGASTREA, INDICATORS OF HOW PARASITISM HAS EVOLVED IN THE NEO~E~~TA? Each group of animals normally shows a mixture of archaic (plesiomorphic) and newly evolved (apomorphic) characters. Applying this to the

1106

K. ROHDE

Fig. 4. Scanning electron micrographs of Lobatostoma manteri {Aspidogast~a). (A) Whole mount, ventral view. (B) Some alveoles of the adhesive disc at higher magnification showing many sensory receptors (S). a - adhesive Fig. 5. Nervous system of larval Muiticofyle pztrvisi disc. h - head. Scale bars 0.5 Frn (A), 50 (.tm (B). (Aspidogastrea), dorsal view. Reconstructed from serial Aspidogastrea, it is impossible to say which of the sections stained with urea-silver nitrate. Modified from Rohde (1968). characters of the group is ancient and was present already when the first neodermatans invaded their hosts. Nevertheless, a comparison of aspidogastreans (and of some other neodermatans) with free-

manteri, which remains enclosed in the egg until it is

living piatyhelminths may permit some conclusions concerning likely characteristics of the first Neodermata. First, parasitism has not led to a reduction in the complexity of the nervous system and sensory apparatus (“sacculinization”), as believed by many. Figures 5-7 show the great complexity of the nervous system of an aspidogastrid and Rohde (1970) even demonstrated the presence of a sheath around parts of posterior nerves in adult Multicotyle purvisi, closely resembl~g a myelin sheath in vertebrates. A comparison of the data in Table 1 makes it clear that the number of receptor types found in parasitic species is not generally smaller than those in free-living ones. The greatest number of receptor in a larval aspidogastrid, types was found ~ulticotyle purvisi, the larva of which swims in freshwater and infects a snail (Figs 8 and 9). In

eaten by a snail, a considerably smaller number of receptor types is present, indicating that the complexity of the sensory apparatus corresponds to the complexity of stimuli to which the larva must react in order to infect a host. Even endoparasitic juvenile and adult aspidogastrids, as well as other endoparasitic neodermatans, possess a great variety of receptors, generally not smaller than that of freeliving turbellarians, and the number of receptors of all types is enormous, reaching 20,000 to 40,000 in juvenile L~batost~~~, endoparasitic in marine snails (Rohde, 1989; Rohde & Watson, 1989). Thus, it seems likely that the first neodermatan was a complex organism possessing a complex nervous system and a great variety of sensory receptors, enabling it to find a host and survive in it (see also Brooks & McLennon, 1993). Secondly, for several aspidogastreans a long

the larva

of another

aspidogastrid,

~butosto~u

1107

Origins of parasitism in Platyhelmin~es II

Fig. 6. Anterior part of the nervous system of adult ikfulticotyle purvisi (Aspidogastrea), reconstructed from serial sections stained with urea-silver nitrate. Modified from Rohde (1968).

Fig. 7. Nervous system of adult ~~~~ic~~ylepurvisi (Aspidogastrea) in middle part of body, reconstructed from serial sections stained with urea-silver nitrate. Modified from Rohde (1968). INT - intestine, TES - testis, VIT - vitellaria, VR - vitelline reservoir.

1108 Table 1 -

K. ROHDE Numbers of types of sensory receptors recorded in various free-living and parasitic Platyhelminthes. Based on “sensograms” (descriptions of sets of sensory receptors) published by various authors

Taxon

Habitat

No. of receptor types

Authors

Diopisthoporus Gymnopharyngeus

free-living, marine

9 (plus statocyst)

Smith & Tyler, 1985, 1986

Luridae gen. et sp. nov, (Dalyelliida)

free-living, marine

2 9 (plus statocyst)

Rohde & Watson, 1993a

Monocelididae, Minoninae gen. sp., (Proseriata)

free-living, marine

8-11 (plus eye and statocyst)

Rohde & Watson, unpublished

Schistosoma mansoni,

free-swimming, freshwater

6

Pan, 1980

free-swimming, freshwater

8

Zdarska, 1992

enclosed in egg, endoparasitic

9

larva (Aspidogastrea)

Rohde, 1989; Rohde & Watson, 1992

Lobatostoma manteri,

endoparasitic

8 (+6?)

Rohde, 1989; Rohde & Watson, 1989

larva (Aspidogastrea)

free-swimming, freshwater, endoparasitic

13 (plus eye)

Rohde & Watson, 199Oc,d. e, 1991

Multicotyle purvisi,

endoparasitic

7 (+2?)

Rohde, 1990b

Gyrocotyle urna, larva (Gyrocotylidea)

free-swimming, marine, endoparasitic 0)

8

Xylander, 1984, 1987

Gyrocotyle urna,

endoparasitic

10

Xylander, 1992

free-swimming, freshwater

6(+3?)

Rohde & Garlick 1985a, b, c; Rohde, Watson & Garlick 1986

(Acoela)

miracidium (Digenea) Echinostoma revolutum,

cercaria (Digenea) Lobatostoma manteri,

juvenile (Aspidogastrea) Multicotyle purvisi,

adult (Aspidogastrea)

adult (Gyrocotylidea) Austramphilina elongata,

larva (Amphilinidea)

survival time outside a host in simple media has been demonstrated. For example, juvenile Lobatostoma munteri from marine snails could be kept alive in dilute sea water for up to 52 days (Rohde, 1973, review 1994). Also indicative of a weak dependence on their hosts is the remarkable lack of host specificity, shown for many aspidogastreans (see reviews by Rohde, 1972, 1994). Examples are Lobastostoma ringens, which was shown to infect 14 teleost species of 8 families in the northern Gulf of Mexico (Hendrix & Overstreet, 1977), and Aspidoguster conchicolu which was recorded from 9 of 10 bivalve species in Kentucky Lake, U.S.A. (Duobinis-

Gray, Urban, Sickel, Owen & Maddox, 1991), and from 7 of 13 bivalve species in North America by Huehner & Etges (1981). A loose association with hosts could have developed secondarily from taxa more closely adapted to parasitism, and the possibility cannot be ruled out that extant aspidogastreans show little host-specificity only because the more specific ones did not survive the many evolutionary “accidents” throughout their long history, whereas less specific ones survived better. However, this may seem unlikely in view of the facts that the Aspidogastrea must be considered “primitive” forms on the basis of morphology and DNA sequences (see

Origins of parasitism in Platyhelminthes

lum

Fig. 8. Larval Multicotyk purvisi (Aspidogastrea). Epidermal sensory receptors and (botton right) horizontal section through eye spot. Modified from Rohde & Watson (199Oc, d, 1991).

above) and that a loose association would be expected in species just evolving towards parasitism. Thus, the first neodermatan probably was loosely associated with its hosts, as indicated by a low host specificity and the ability to survive outside a host for long periods. Thirdly, the original hosts of aspidogastreans very likely were molluscs, as indicated by the observation that some species can complete their life cycle in molluscs (e.g. Aspidogaster conchicola, see reviews by Rohde, 1972, 1994, Fig. 10). Vertebrates serve as faculative hosts for such species, in which they can

survive and reproduce when eaten with their mollusc host. In other species, the vertebrate has become an obligatory, final host, where the parasite reproduces (e.g. Lobatostoma mnnteri, see review by Rohde, 1994; Fig. 11). But even in Lobatostoma only the very last stages (maturation of the gonads and reproduction) has been transferred to the vertebrate, and little (if any) active growth occurs at least in some species. A slight increase in body size in L. manteri in the vertebrate host appears to be entirely due to the accumulation of eggs in the uterus. Although the possibility cannot completely be excluded that verte-

Ill0

K. ROHDE

Fig. 9. Larval ~~Ztjcoty~e purvisi (Aspidogastre~). (1) Diagram of larva with two anterior receptor complexes {arrowheads). (2) One of the two anterior receptor complexes. modified from Rohde & Watson (1990e).

Fig. 10. Life cycle of Aspidogmter conchicota (Aspidogastrea). Note: the life cycle can be completed in a single mollusc host, vertebrates may be incorporated as facultative hosts. were parasitized before the molluscs, this seems less likely than the reverse sequence, considering the apparently few factors contributed to completion of the life cycle even by those vertebrate hosts that are obligatory. Experimental studies attempting to bring life cycles to completion in the mollusc might show that only a single hormone is brates

required. Thus, the first neode~atan probably was a parasite of molluscs, although it must be emphasized that, to date, life cycles are known only from species of the family Aspidogastridae. Immature specimens of ~tic~ocoty~e ~e~~~~~~~ (family Stichocotylidae) were found encysted in the intestinal wall of marine crustaceans (references in Rohde,

Origins of parasitism in Platyhelminthes

1111

Fig. 11. Life cycle of Lobatostoma manteri (Aspidogastrea). Adult worms live in the intestine of the teleost fish Trachinotus blochi, eggs are shed in the faeces and are eaten by prosobranch snails, in which juvenile worms reaching (almost) adult size develop. Fish become infected by eating the snails.

1972). I suspect that molluscs serve as first hosts for this species as well. However, if crustaceans are such hosts, the possibility should be considered that the first neodermatans infected arthropods as well as molluscs, and possibly other invertebrates. Further evolution of Neodermata led to a closer incorporation of the vertebrate host in the life cycles of digenean trematodes. Since infection of new hosts is usually (and probably always) dependent on availability of many infective stages in order to compensate for the loss of most of them, digeneans increased their offspring by intercalating multiplicative larval stages (sporocysts, rediae) in the mollusc host. Cercarial dispersal stages were “invented” to secure entrance into the vertebrate. Evolution of the life cycles of Monogenea and Cestoda is unclear. Ultrastructure of the protonephridia (and other characteristics) show that monogeneans are most closely related to the trema-

todes (e.g. Rohde, 1993); they may have lost the primitive mollusc host secondarily “specializing” on the vertebrate host, or they may have split off from the trematode line very early, before the latter had become parasitic in molluscs. Cestodes are the most derived of all Neodermata, as indicated by DNA studies (Blair, 1993), and by a protonephridial system lacking junctions in the proximal capillaries and flame bulbs, and possessing short microvilli in the capillaries instead of lamellae found in trematodes and monogeneans, clearly derived (apomorphic) features (Rohde, 1993; for further details see Ehlers, 1985) (Fig. 3). Their life cycles incorporate arthropod intermediate and vertebrate final hosts. As in the Monogenea, it is unclear how their life cycle has evolved. They either replaced the original mollusc with an arthropod host or split off the “primitive” neodermatan line early, before molluscs had been incorporated in the life cycle. Alternatively,

1112

K. ROHDE

they may have retained an original arthropod host used by some early neodermatans (see discussion on Siic~oeoi~le above). Cues for the geological time span of platyheiminth evolution can be found in the types of host used by the aspidogastreans and digeneans. Gibson (1987) discussed the time in geological history when digeneans first invaded their hosts and concluded that acquired their mollusc hosts “proto-trematodes” “somewhere around the Silurian period” about 400 million years ago, when Cercomeromorphae (monogeneans and cestodes) aiso acquired their hosts, the first vertebrates, i.e. placoderms. According to Gibson, an even earlier origin of trematodes is possible but not likely. Vertebrate hosts were acquired by trematodes much later, “probably not until as late as the Triassic period c.200 million years ago”. Aspidogastreans belonging to the Stichocotylida (system of Gibson & Chinabut, 1984) are parasitic in elasmobranchs: holocephalans and selachians, whereas Aspidogastrida occur in teleosts and reptiles. Digenea are almost exclusively parasites of teleosts and “higher” vertebrates, but not of elasmobranchs. According to Young (1950) and Colbert (1980), true elasmobranchs first appeared in the middle/late Devonian, presumably derived from some placodenn perhaps in the Ordovician. They split into holocephalans and selachians probably in the Carboniferous. Teleosts first appeared in the Jurassic, derived from holosteans. Distribution of the aspidogastreans in different host groups suggests that Aspidogastrea invaded their vertebrate hosts in the Devonian or earlier, Aspidogastridae in the Jurassic. AN ALTERNATIVE HYPOTHESIS: MOSAIC EVOLUTION? Recent DNA studies of various organisms

have

shown that, at least in certain groups, horizontal (= lateral) gene transfer between species is possible (Margulis & Fester, 1991; Cummings, 1994). Rohde (1990a) discussed the implications of such transfer for the evolution of eukaryotes and particularly platyhelminths and concluded that, if horizontal gene transfer of evolutionary significance should occur (even if only rarely), it might have profound impact on the phylogeny of many groups and lead to what he called “mosaic evolution”. Organisms

“would not evolve as a whole but - to a certain degree - the units of evolution would be character traits, exchangeabie between phylogenetic lines”. It would then be impossible to establish definitive phylogenetic trees for particular groups. Appiied to the platyhelminths discussed in this paper, conclu-

sions concerning phylogenetic relationships would not apply to all the characte~stics of the taxa, but only to some obvious, major features. In other words, Aspidogastrea may be the closest relatives of the Digenea in most respects, but closest relatives of some other taxa in others, and the Neodermata may be manophyletic only in the major characteristics. Acknowledgements -

Financial support was given by the University of New England and the Australian Research Council. The electron-microscopic work was done jointly with Nikki Watson. I thank Nikki Watson for critically reading the manuscript and for bibliographic assistance. Becky Francis redrew some of the figures. REFE~N~ES Anderson T. J., Newman L. J. & Lester R. J. G. 1993. Light and electron microscopic study of Urosporidium cannoni n.sp., a haplosporidian parasite of the polyclad turbellarian Stylochus sp. Journal of Eukaryotic Microbiology 40: 162-168.

Baverstock P. R., Fielke R., Johnson A. M., Bray R. A. & Beveridge I. 1991. Conflicting phylogenetic hypotheses for the parasitic Platyhelminthes tested by partial sequencing of 18s ribosomal RNA. International Journal for Parasitology 21: 329-339.

Blair D. & Williams J. B. 1987. A new fecampiid of the Kronborg~a ~latyleminthes: T’urbeIlaria: genus Neorhabd~~la) parasitic in the intertidal isopod ~~osphaeroma obtusum (Dana) from New Zealand. Journal of Natural History 21: 1155-1172.

Blair D. 1993. The phylogenetic position of the Aspidobothrea within the parasitic flatworms inferred from ribosomal RNA sequence data. International Journal for Parasitology 23: 169-178.

Bresciani J. & Koie M. 1970. On the ultrastructure of the epidermis of the adult female of Kronborgia amphipodicola Christensen and Kanneworff, 1964 (Turbellaria: Neorhabdocoela). Gpheha 8: 209-230. Brooks D. R. 1989. A summary of the database pertaining to the phylogeny of the major groups of parasitic platyhe~inths, with a revised classification. Can~~an Journal 0~200~0gy 67: 714-720.

Brooks D. R. & McLennan D. A. 1993. Macroevolutionary patterns of morpholigical diversification among parasitic flatworms (Platyhelminthes: Cercomeria). Evolution 47: 495-509. Burt M. D. B. 1988. Recent advances in our knowledge of parasitic Turbellaria. Fortschritte der Zoologie 36: 15-23. Cannon L. R. G. 1982. Endosymbiotic umagillids (Turbellaria) from holothurians of the Great Barrier Reef. Zoologica Scripta 11: 173-188. Cannon L. R. G. 1986. Turbellaria of the World. A Guide to Families and Genera. Queensland Museum, Brisbane. Cannon L. R. G. 1987. Two new rhabdocoel turbella~ans, endosymbiotic with holothu~ans (~hinode~ata) from the Great Barrier Reef; and a discussion of sclerotic structures in the female system of the Umagillidae. Zoologica Scripta 16:297-303.

Origins of parasitism in Platyhelminthes Cannon L. R. G. & Jennings J. B. 1987. Occurrence and nutritional relationships of four ~tos~bionts of the freshwater crayfishes Cherux d&pm Riek and Cherax punctatus Clark (Crustacea: Decapoda) in Queensland. Australian Journal of Marine and Freshwater Research 1987:419427. Colbert E. H. 1980. Evolution of the Vertebrates. 3rd edn.

John Wiley, New York. Cummings M. P. 1994. Transmission patterns of eukaryotic transposable elements: arguments for and against horizontal transfer. Trends in Ecology and Evolution 9: 141-145.

Duobinis-Gray L. F., Urban E. A., Sickel J. B., Owen D. A. & Maddox W. E. 1991. Aspidogastrid (Trematoda) parasites of unionid (Bivalvia) molluscs in Kentucky Lake. Journal of the ~e~~~ho~og~eaZ Society of Washington 58: 167-l 70.

Vereins Hamburg 27: 29 l-294. U. 1985. Ilas Phyiogene~~ehe System der Plathebninthes. Gustav Fischer, Stuttgart, New York.

Ehlers

Fleming L. C. 1986. Occurrence of symbiotic turbellarians in the oyster, Crassostrea virginica. Hydrobiologia 132: 311-315.

Gibson D. I. & Chinabut S. 1984. Rohdella siumensis gen. et sp, nov. (Aspidogastridae: Rohdellinae subfam. nov.) from freshwater fishes in Thailand, with a reorganization of the classification of the subclass Aspidogastrea. ParasitoZogy 88: 383-393.

Gibson D. I. 1987. Questions in digenean systematics and evolution. Parasitology 95: 429-460. Hendrix S. S. & Overstreet R. M. 1977. Marine aspidogas&ids (Trematoda) from fishes in the northern Gulf of Mexico. Journal of Parasitology 63: 8 10-S 17. Holt P. A. & Mettrick D. F. 1975. Ultrastructural studies of the epidermis and gastrodermis of Syndesmisfranciscana (Turbellaria: Rhabdocoela). Canadian Journal of Zoology 53: 536-549.

M. K. & Etges F. J. 1981. Encapsulation

B.), pp. 127-160. University of South Carolina Press, Columbia, South Carolina. Jennings J. B. 1977. Patterns of nutritional physiology in free-living and symbiotic Turbellaria and their implications for the evolution of entoparasitism in the phylum Platyhe~inthes. Acta Zooiogicu Fen&a 154: 63-79. Jennings J. B. Br LeFlore W. B. 1979. Occurrence of possible adaptive significance of some histochemically demonstrable dehydrogenases in two entosymbiotic rhabdocoels (Platyhelminthes: Turbellaria). Comparative Biochemistry and Physiology Part B 62231 301-304.

Jennings J. B. 1980. Nutrition in symbiotic Turbellaria. In: Nutrition in the Lower Metazoa (Edited by Smith D. C. & Tiffon Y.), pp. 45-56. Pergamon Press, Oxford. Jennings 3. B. 1981. Physiological adaptations to entosymbiosis in three species of graffillid rhabdocoels. Hydrobiologia 84: 147-153.

Ehlers U. 1984. Phylogenetisches System der Plathelminthes. Verhandhmgen des Na~rw~sensc~ftli~hen

Huehner

1113

of

Aspidogaster conchicoia (Trematoda: Aspidogastrea) by unionid mussels. Journal of Invertebrare Pafhology 37: 123-128. Hyra G. S. 1993. Genostoma kozlosfi sp. nov. and G. inopinatum sp. nov. (Turbellaria: Neorhabdocoela:

Genostomatidae) from leptostracan crustaceans of the genus Nebaha. Cahiers de Siofogie Marine 34: 11l-126. Jennings J. B. & Mettrick D. F. 1968. Observations on the ecology, morphology and nutrition of the rhabdocoel turbellarian Syndesmis franciscana (Lehman, 1946) in Jamaica. Caribbean Journal of Science 8: 57-69. Jennings J. B. 1971. Parasitism and commensalism in the Turbellaria. Advances in Parasitology 9: l-32. Jennings J. B. 1974a. Digestive physiology of the Turbellaria. In: Biology of the Turbellaria (Edited by Riser N. W. & Morse M. P.). pp. 173-197. McGrawHill, New York. Jennings J. B. 1974b. Symbioses in the Turbdlaria and their implications in studies on the evolution of parasitism. In: Symbiosis in the Sea (Edited by Vernberg W.

Jennings J. B. & Cannon L. R. G. 1987. The occurrence, spectral properties and probable role of haemoglobins in four entos~biotic tur~~a~ans (Rhabd~~la: Umagilhdae). Opkelia 27: 143-154. Jennings J. B. 1988. Nutrition and respiration in symbiotic Turbellaria. Fortschritte der Zoologie 36: l-13. Jennings J. B. 1989. Epidermal uptake of nutrients in an unusual tur~lla~an parasitic in the starfish Coscinasferias calamaria in Tasmanian waters. Biological Bulletin 176: 327-336.

Jennings J. B. & Hick A. J. 1990. Differences in the distribution, mitochond~al content and probable roles of haemo~obin~ontaining parench~al cells in four species of entosymbiotic turbellarians (Rhabdocoela: Umagillidae and Pterastericolidae). Opheha 31: 163-175.

Joffe B. I. 1987. [On the evolution of the pharynx in the Platyhelminthes]. In: [Morphology of Turbellarians, Pogonophores and Ascidians] (Edited by Mamkaev Yu. V.), pp. 34-71. USSR Academy of Sciences, Proceedings of the Zoological Institute, 167 (in Russian). Joffe B. I., Slusarev G, S. & Timofeeva T. A. 1987. [Pharynx structure in the monogeneans and their phylogenetic relationship with the turbellarians.] Parazitoiogiya 21: 472-481 (in Russian). Joffe B. I. 1988. On the phylogeny of Temn~phalida. Fortschritte der Zoologie 36: 59-62.

Joffe B. I. & Chubrik G. K. 1988. [The structure of the pharynx in trematodes and phylogenetic relations between Trematoda and Turbellaria]. Para~jto~ogjya 22: 297-303 (in Russian). Jondelius U. 1986. Epidermal ultrastructure of Anoplodium stichopi (Plathelminthes, Dalyellioida), a flatworm endosymbiotic in Stichopus tremulus (Holothurida). Zoomorphology 106: 254259.

Jondelius U. 1989. Pterastericula pellucjda n. sp., an endosymbiotic flatworm (Dalyellioida: Pterastericolidae) from Astropecten irregularis, with notes on the monophyly of the genus P~eras~ericola. Ophelia 30: 55-62. Jondelius U. 1992. Adhesive glands in the ~erastericolidae (Plathelminthes, Rhabdocoela). Zoomorphology 111: 229-238.

Jondelius U. & Thollesson

M. 1993. Phylogeny of the

1114

K. ROHDE

Rhabdocoela

(Platyhelminthes):

a working hypothesis.

Canadian Journal of Zoology 71: 298-308.

Jones T. C. & Lester R. J. G. 1992. The life history and biology of Dicerafocephala boschmai (Platyhelminthes; Temnocephalida), an ectosymbiont on the redclaw crayfish Cherax quadricarinatus. Hydrobiologia 248: 193-199. Justine J.-L. 1991. Phylogeny of parasitic Platyhelminthes: a critical study of synapomorphies proposed from the ultrastructure of spermiogenesis and spermatozoa. Canadian Journal of Zoology

69: 1421-1440.

B. & Christensen A. M. 1966. Kronborgia caridicola sp. nov., an endoparasitic turbellarian from north Atlantic shrimps. Ophelia 3: 65-80.

Kanneworff

Keie M. & Bresciani J. 1973. On the ultrastructure of the larva of Kronborgia amphipodicola Christensen and Kanneworff 1964 (Turbellaria, Neorhabdocoela). Ophelia 12: 171-203.

Margulis L. & Fester R. (Ed.) 1991. Symbiosis as a Source of Evolutionary Innovation: Speciation genesis. MIT Press, Cambridge.

and Morpho-

Mettrick D. F. & Jennings D. B. 1969. Nutrition and chemical composition of the rhabdocoel turbellarian Syndesmis franciscana, with notes on the taxonomy of S. antillarum. Journal of the Fisheries Research Board of Canada 26: 2669-2679.

Noury-Srdiri N., Justine J.-L. & Euzet L. 1989. Ultrastructure du ttgument de trois especes de Paravortex (Rhabdocoela, “Dalyellioida”, Graffillidae), turbellaries parasites intestinaux de mollusques. Annales des Sciences NaturelIes, Zoologie (13 Zme Serie) 10: 155-170. Ott J., Rieger G., Rieger R. & Enders F. 1982. New mouthless interstitial worms from the sulfide system: symbiosis with prokaryotes. Pubblicazioni della Stazzione Zoologica di Napoli I: Marine Ecology 3: 313-333.

Pan S. C.-T. 1980. The fine structure of the miracidium of Schistosoma mansoni. Journal of Invertebrate Pathology 36: 307-372.

Popova N. V. KcMamkaev Y. V. 1985. [Ultrastructure and primitive features of the eyes of Convoluta convoluta (Turbellaria, Acoela)]. Doklady Akademii Nauk SSSR 283: 756759.

Rohde K. 1968. The nervous systems of Multicotyle purvisi Dawes, 1941 (Aspidogastrea) and Diaschistorchis multitesticularis Rohde, 1962 (Digenea). Implications for the ecology of the parasites. Zeitschrtft fir Parasitenkunde 30: 78-94.

Rohde K. 1970. Nerve sheath in Multicotylepurvisi Dawes. Naturwissenschaften 57: 502-503.

Rohde

K.

1971. Phylogenetic

origin

of

trematodes.

Parasitologische Schriftenreihe 21: 17-27.

Rohde K. 1972. The Aspidogastrea, especially Multicotyle purvisi Dawes, 1941. Advances in Parasitology 10: 77-151. Rohde K. 1973. Structure and development of Lobatostoma manteri nov. (Trematoda: SP. Aspidogastrea) from the Great Barrier Reef, Australia Parasitology 66: 63-83.

Rohde K. 1976. Species diversity of parasites on the Great Barrier Reef. Zeitschrift fur Parasitenkunde 50: 93-94. Rohde K. & Garlick P. R. 1985a. Two ciliate sense recep-

tors in the larva of Austramphilina elongata Johnston, 193, (Amphilinidea). Zoomorphology 105: 30-33. Rohde K. & Garlick P. R. 1985b. Subsurface sense receptors in the larva of Austramphilina elongata Johnston, 1931 (Amphilinidea). Zoomorphology 105: 34-38. Rohde K. & Garlick P. R. 1985~. Ultrastructure of the posterior sense receptor of larval Austramphilina elongata Johnston, 1931 (Amphilinidea). International Journal for Parasitology 15: 399-402.

Rohde K., Watson N. & Garlick P. R. 1986. Ultrastructure of three types of sense receptors of larval Austramphilina elongata (Amphilindea). International Journal for Parasitology 16: 245-25 1.

Rohde K. 1988. Phylogenetic relationship of free-living and parasitic Platyhehninthes on the basis of ultrastructural evidence. Fortschritte der Zoologie 36: 353-357. Rohde K. & Watson N. 1988. Ultrastructure of epidermis and sperm of the turbellarian Syndisyrinx punicea (Hickman, 1956) Umagillidae). (Rhabdocoela, Australian Journal of Zoology 36: 131-139. Rohde K., Watson N. & Cannon L. R. G. 1988. Comparative ultrastructural studies of the epidermis, sperm and nerve libres of Cleistogamia longicirrus and Seritia stichopi (Rhabdocoela, Umagillidae). Zoologica Scripta 17: 337-345.

Rohde K. 1989. At least eight types of sense receptors in an endoparasitic flatworm: a counter-trend to sacculinization. Naturwissenschaften 76: 383-385. Rohde K. & Watson N. 1989. Sense receptors in Lobatostoma manteri (Trematoda, Aspidogastrea). International Journal for Parasitology 19: 847-858.

Rohde K. 1990a. Phylogeny of Platyleminthes, with special reference to parasitic groups. International Journal for Parasitology 20: 979-1007.

Rohde K. 1990b. Ultrastructure of the sense receptors of adult Multicotyle purvisi (Trematoda, Aspidogastrea). Zoologica Scripta 19: 233-241.

Rohde K. & Watson N. 1990a. Epidennal and subepidermal structures in Didymorchis (Platyhelminthes, Rhabdocoela) I. Ultrastructure of epidermis and subepidermal cells. Zoologischer Anzeiger 224: 263-275. Rohde K. & Watson N. 1990b. Epidermal and subepidermal structures in Didymorchis (Platyhelminthes, Rhabdocoela) II. Ultrastructure of gland cells and ducts. Zoologischer Anzeiger 224: 276-285.

Rohde K. & Watson N. A. 1990~. Non-ciliate sensory receptors of larval Multicotyle purvisi (Trematoda, Aspidogastrea). Parasitology Research 76: 585-590. Rohde K. Watson N. A. 1990d. Uniciliate sensory receptors of larval Multicotyle purvisi (Trematoda, Aspidogastrea). Parasitology Research 76: 591-596. Rohde K. & Watson N. 1990e. Paired multiciliate receptor complexes in larval Multicotyle purvisi (Trematoda, Aspidogastrea). Parasitology Research 76: 597601. Rohde K. 1991. The evolution of protonephridia of the Platyhelminthes. Hydrobiologia 227: 3 15-321. Rohde K. & Watson N. 1991. Ultrastructure of pigmented photoreceptors of larval Multicotyle purvisi (Trematoda, Aspidogastrea). Parasitology Research 77: 485-490. Rohde K. & Watson N. A. 1992. Sense receptors of larval

Origins of parasitism in Platyhelminthes Lobatostoma manteri (Trematoda, Aspidogastrea). International Journal for Parasitology 22: 3542.

Rohde K. 1993. Ultrastructure of protonephridia in the Monogenea. Implications for the phylgeny of the group. Bulletin Francais de la P&he et de la Pisciculture 328: 115-119. Rohde K., Hefford C., Ellis J., Baverstock P. R., Johnson A. M., Watson N. A. & Dittmann S. 1993. Contributions to the phylogeny of Platyhelminthes based on partial sequencing of 18s ribosomal DNA. International Journal for Parasitology 23: 705-724.

Rohde K. & Watson N. A. 1993a. Ultrastructure of sensory receptors of an undescribed species of Luridae (Platyhelminthes: Rhabdocoela). Australian Journal of Zoology 41: 53-65.

Rohde K. & Watson N. A. 1993b. Spermatogenesis in Udonella (Platyhelminthes, Udonellidea) and the phylogenetic position of the genus. International Journal for Parasitology 23: 1255135.

Rohde K., Luton K, Baverstock P. R. & Johnson A. M. 1994. The phylogenetic relationships of Kronborgia (Platyhelminthes, Fecampiidae) based on comparison of 18s ribosomal DNA sequences. International Journal for Parasitology 24: 657-669.

Rohde K. 1994. The minor groups of parasitic Platyhehninthes. Advances in Parasitology 33: 145-234. Saffo M. B. 1992. Invertebrates in endosymbiotic associations. American Zoologist 32: 557-565. Shinn G. L. 1981. The diet of three species of umagillid neorhabdocoel turbellarians inhabiting the intestine of echinoids. Hydrobiologia 84: 155-162. Shinn G. L. 1985a. Reproduction of Anoplodium hymanae, a turbellarian flatworm (Neorhabdocoela, Umagillidae) inhabiting the coelom of sea cucumbers; production of egg capsules and escape of infective stages without evisceration of the host. Biological Bulletin 169: 182-198. Shinn G. L. 1985b. Infection of new hosts by Anoplodium hymanae, a turbellarian flatworm (Neorhabdocoela, Umagillidae) inhabiting the coelom of the sea cucumber

1115

(Turbellaria: Acoela). Journal of the Marine Biological Association of the United Kingdon 51: 301-313. Watson N., Rohde K. & Williams J. B. 1992. Ultrastructure of the protonephridial system of larval Kronborgia isopodicola (Platyhelminthes, Fecampiidae). Journal of Submicroscopic 4349.

Cytology and Pathology 24:

Watson N. A., Williams J. B. & Rohde K. 1992. Ultrastructure and development of the eyes of larval Kronborgia isopodicola (Platyhelminthes, Fecampiidae). Acta Zoologica (Stockholm) 73: 95-102.

Watson N. A. & Rohde K. 1993a. Ultrastructural evidence for an adelphotaxon (sister group) to the Neodermata (Platyhelminthes). International Journal for Parasitology 23: 285-289.

Watson N. A. & Rohde K. 1993b. Ultrastructure of sperm and spermiogenesis of Kronborgia isopodicola (Platyhelminthes, Fecampiidae). International Journalfor Parasitology 23: 131-744.

Williams J. B. 1988. Ultrastructural (Platyhelminthes, Neophora):

studies on Kronborgia the spermatozoon.

International Journal for Parasitology 18: 417483.

Williams J. B. 1990a. Ultrastructural studies on Kronborgia (Platyhelminthes: Fecampiidae): epidermis and subepidermal tissue of the parasitic male K. isopodicola. International Journal for Parasitology 20: 329-340.

Williams J. B. 1990b. Ultrastructural studies on Kronborgia (Platyhelminthes: Fecampiidae): subepidermal glands of the female K. isopodicola. International Journal for Parasitology 20: 803-8 13.

Stichopus californicus. Biological Bulletin 169: 199-214.

Williams J. B. 1991. Ultrastructural studies on Kronborgia (Platyhelminthes: Fecampiidae): observations on the encapsulated larva of K. isopodicola. New Zealand Journal of Zoology 18: 251-265. Xylander W. E. R. 1984. A presumptive ciliary photoreceptor in larval Gyrocotyle urna Grube and Wagener (Cestoda). Zoomorphology 104: 21-25. Xylander W. E. R. 1987. Ultrastructure of the lycophora larva of Gyrocotyle urna (Cestoda, Gyrocotylidea). II. Receptors and nervous system. Zoologischer Anzeiger

Smith J. P. S. & Tyler S. 1985. Fine structure and evolutionary implications of the frontal organ in Turbellaria Acoela. I. Diopisthoporus gymnopharyngeus sp. n.

Xylander W. E. R. 1992. Sinneszellen von Gyrocotyle urna: Rezeptorenvielfalt bei einem urspriinglichen Cestoden.

Zoologica Scripta 14: 91-102.

Smith III J. P. S. & Tyler S. 1986. Frontal organs in the Acoelomorpha (Turbellaria): ultrastructure and phylogenetic significance. Hydrobiologia 132: 71-78. Stunkard H. W. 1917. Studies on North American Polystomidae, Aspidogastridae, and Paramphistomidae. Illinois Biological Monographs 122: 137-148.

Stunkard H. W. 1967. Platyhelminthic parasites of invertebrates. Journal of Parasitology 53: 673-682. Taylor D. L. 1971. On the symbiosis between Amphidinium klebsii (Dinophyceae) and Amphiscolops langerhansi

219: 239-255.

Verhandlungen der Deutschen Zoologischen 85: 230.

Gesellschaft

Yamasu T. 1991. Fine structure and function of ocelli and sagittocysts of acoel flatworms. Hydrobiologia 227: 273-282.

Young J. Z. 1950. The Life of Vertebrates. Clarendon Press, Oxford. Zdarska Z. 1992. Transmission electron microscopy of sensory receptors of Echinostoma revolutum (Froelich 1802) cercaria (Digenea: Echinostomatidae). Parasitology Research 78: 598-606.