Representative worms from the three major monophyletic groups within the phylum Platyhelminthes: the Acoelomorpha (represented by an acoel turbellarian), the Catenulida (represented by a retronectid turbellarian), and the Rhabditophora (represented by a digenetic trematode)

The relatively small worms that are commonly called flatworms are classified in the phylum Platyhelminthes, whose Greek roots mean, in fact, "flat worm." In general, the phylum encompasses worms that

• are softbodied (lacking cuticle or other rigid skeletal support),
• have a solid body form (lacking the fluid-filled space, such as coelom, pseudocoel, or hemocoel, between body wall and gut that is seen in other animals),
• are hermaphroditic (individuals have both male and female reproductive organs and mutually inseminate in sexual reproduction),
• have spermatozoa with two tails (a pair of flagella arise from the paired centrioles of the developing sperm cell),
• have a sack-like gut (the only opening to the digestive tissue being the mouth),
• have epidermal cells that are typically multiciliated, each cell bearing multiple cilia rather than only one, and
• have fiber-form muscles (cells residing within the mesenchyme, without evidence of epithelial origin).

They thus stand distinct from the major phylum of worms, the Annelida, whose members have a complete gut (with anus as well as mouth), segmented bodies with fluid-filled coelomic compartments, a cuticle-covered body wall, and muscles that arise from epithelial mesodermal tissue, and monoflagellated sperm (as do most other animals). But other phyla of small worms share some of the characters that otherwise set platyhelminths apart. Hermaphroditism, with reproductive organs as complicated as those of the flatworms, appear in the Gnathostomulida and Gastrotricha, for instance. Lack of a cuticle covering the body wall is also a feature of the Gnathostomulida (though the jaws of gnathostomulids are true cuticular elements which are completely lacking in platyhelminths), and so, too, is a sack-like gut a feature of the gnathostomulids (but it appears to be derived from a complete gut by virtue of evidence for a vestigial anus in gnathostomulids). Other worm phyla have cuticle (Gastrotricha, Nematoda, Nematomorpha, Kinorhyncha, Priapulida, etc.), multiciliated epidermal cells (all worm phyla but the Gnathostomulida), solid (acoelomate) bodies (Gastrotricha and miscellaneous representatives of Nematoda, Annelida, etc.), and fiber-form muscle cells. And though obviously a convergent similarity, the character of biflagellate spermatozoa is seen in some groups of fishes.

Nevertheless, the characteristics we can list for the phylum Platyhelminthes are not absolute distinctions--none apply to all groups together to the exclusion of other animals. Because no overarching characteristics (synapomorphies) for the phylum as a whole can be identified unequivocally, there is ground for considering the Platyhelminthes not to be a valid monophyletic phylum (Smith et al., 1986). At the moment, the largest monophyletic groups that can be identified among flatworms are three, the Acoelomorpha, the Catenulida, and the Rhabditophora, and while any two of these may be related as sister groups, the characters we could use to tie those two would exclude the third from falling into a logical phylogenetic relation with them. For example, the Catenulida and Rhabditophora appear to share homologies in the structure of the epidermal ciliary rootlets, in the mechanism by which they replace their epidermal cells, and in having protonephridia, but these homologies do not extend to the Acoelomorpha. The Acoela, on the other hand, appears to share with the Rhabditophora the homology of biflagellate spermatozoa, but since catenulids don't have such sperm, this homology would negate those homologies listed between Catenulida and Rhabditophora. The Acoela shares no apomorphies with both the Catenulida and Rhabditophora except possibly the presence of neoblasts (Rieger and Ladurner, 2001).

Turbellarian platyhelminths; representatives of 4 of the 11 orders of turbellarians

The current classification system for the Platyhelminthes is a cladistic one recognizing these three monophyletic groups and dispensing with the traditional division of the phylum into the four classes Turbellaria, Monogenea, Trematoda, and Cestoda. Turbellaria is, in particular, considered an invalid class because it is paraphyletic (Ehlers, 1985). (That is, because the parasitic classes arose from an ancestor that would be classified within the Turbellaria, the group Turbellaria is not monophyletic, not all of its descendants being encompassed within it.) Turbellarians are the largely free-living flatworms, those that don't parasitize other animals, while the other classes encompass the obligate parasites. Eleven orders of turbellarians are recognized in the commonly used classification that Hyman championed (Tyler, 1999): Nemertodermatida, Acoela, Catenulida, Haplopharyngida, Macrostomida, Polycladida, Lecithoepitheliata, Prolecithophora, Rhabdocoela, Proseriata, and Tricladida). Monogeneans are largely ectoparasites on vertebrates like fishes; trematodes are the flukes, most of which live inside the organs of vertebrates as adults; and cestodes are the tapeworms, living in the intestines of vertebrates as adults. The term "turbellarian" is still validly applied to the free-living flatworms, but the term "Turbellaria" (that is, the official, capitalized taxon name) should be written in quotation marks to indicate its paraphyletic status. Not all turbellarians are free-living, and so "free-living plathelminths," another name often applied to them by cladists wishing to avoid "Turbellaria," has some disadvantages; there are some highly specialized parasites among virtually all subgroups of the turbellarians.

Turbellarian platyhelminths; representatives of another three orders of turbellarians

The quintessential parasitic flatworms are the monogeneans, trematodes, and cestodes, and these constitute monophyletic groups, but by a cladistic classification, these monophyletic groups would not have the rank of class. Instead, if any monophyletic groups are to be considered classes, they would have to be the three major groups constituting the former "Turbellaria," namely the Acoelomorpha, Catenulida, and Rhabdocoela. The major parasitic groups lie within the Rhabdocoela, specifically within the monophyletic taxon Neodermata in that class. They are clearly closely related, descended from a common ancestor among the turbellarians.

Representatives of the three major groups of Neodermata (a tapeworm, a digenetic trematode, and a monogenean)

Exactly where the Neodermata came from among turbellarian platyhelminths is not at all clear; a number of potential ancestral groups have been proposed. The major advancement that allowed them to adopt a parasitic existence seems to be the neodermis which arises as the parasites attack a new host and metamorphose from the free-living larva. That larva has an epidermis much like that of turbellarians, composed of ciliated cells. When it locates a host, the larva sheds this ciliated epidermis and a new epidermis, the neodermis, emerges from cells situated below the muscle layer of the body wall. These cells fuse to create a syncytial covering over the entire body. The neodermis, thus, is an uninterrupted layer of syncytium whose nuclei lie in cytons below the body-wall musculature; on its apical surface facing the environment are specialized microvilli-like projections whose shape is specialized in each of the neodermate groups. The neodermis must offer advantages in a parasitic existence, allowing the parasite to absorb nutrients from the host (cestodes have, in fact, dispensed with the mouth and gut and gain all their nutrients from the host by absorption through the neodermis) and probably serving a dynamic role in defeating host immune reactions. Platyhelminths seem to be preadapted to developing such an epidermis in that even the turbellarians go through successive generations of epidermis in their embryonic development, and as adults they regenerate their epidermis by repacing cells that are lost with cells that migrate into it from the parenchyma below the muscles (Tyler and Tyler 1996).

---

The Acoela and its relationship to the Platyhelminthes

Convoluta pulchra Smith and Bush 1991, a typical mud-inhabiting acoel that feeds on diatoms

This group of quite small worms is so-named because of their lack of a gut cavity ("A" for non; "coela" for cavity [gut, in this case, not coelom]). Their digestive tissue is a syncytium that encompasses food items in large vacuoles rather than within an epithelially lined lumen as do the guts of other animals. Because no epithelium delimits the gut contents, they appear solid-bodied (acoel). (Some other turbellarian groups are even named according to the shape of their gut--the triclads with three-branched guts, the polyclads with many-branched guts, for instance.) By some older, no longer accepted theories, the syncytial nature of the gut in acoels allied them with the ciliate protozoans or with a hypothetical protozoan-derived phagocytic ancestor, but electron microscopy has shown that this syncytiality arises secondarily from fusion of progenitor cells among so-called wrapping cells in the parenchyma surrounding it. The digestive tissue in some acoels is cellular, with spaces between these cells serving as a sort of lumen but not an epithelium-bounded lumen as in other turbellarians (Smith and Tyler, 1985).

Almost all acoels live in marine habitats, mostly in between the grains of marine sediments but some swim in the plankton or creep on algae and other marine substrates. One feature that is immediately apparent in acoels is the statocyst, a glass-like sphere at the anterior end of the body in the brain, with a capsule surrounding a hemispherical concretion, presumably used to sense the direction of gravity and other acceleration. Also prominent in acoels are parts of the reproductive system, which, as in other flatworms, is a complicated hermaphroditic system with both male and female components. As with many other animal groups, classification of the acoel species is based on the variety of form in these reproductive organs. Like other flatworms, the sperm in acoels are biflagellate (having two tails or flagella).

Aphanostoma bruscai Hooge and Tyler 2002, schematic drawing by Matt Hooge of the internal organs reconstructed from sections

The seemingly primitive and varied nature of the nervous system in many acoels makes them good candidates for models of the ancestral condition of the nervous system among flatworms (Reuter et al., 2001 [17294]). But acoels have some peculiar features that set them off strongly from other flatworms as well as from other phyla of animals. No other animals have such a digestive syncytium instead of an epithelial gut, and the extreme reduction (if not complete absence) of the extracellular matrix is peculiar; acoels have no basement membrane under their epidermis or other epithelia, something that is unknown among any other animals. The cilia of acoels, the locomotory organelles by which they swim, have a unique ultrastructure as well. Unlike cilia of other metazoans, the epidermal cilia of acoels are interconnected by a complicated system of rootlets, and they also have a distinctive shape to their tips--a sort of abrupt narrowing set off from the main part of the cilium by a sharp shelf.

Indications that these peculiar features are secondary specializations come from comparison of acoels with their closest relatives, the Nemertodermatida. Nemertodermatids resemble acoels in many respects and were once classified in a family within the Acoela. Their statocyst is double, however (having two stones within it), and their gut, while not as distinctly sack-shaped as other turbellarians, is epithelially lined (Smith, 1981; Smith and Tyler, 1985). Together the Acoela and Nemertodermatida are sister groups constituting the Acoelomorpha. Their sister-group relationship is strongly supported by similarities in their ciliation (nemertodermatids have their cilia interconnected by ciliary rootlets and have shelfed tips), and similarities in body-wall structure (reduction of extracellular matrix), reproductive organs (lack of sperm ducts and lack of female ducts), and the relation of the statocyst to the nervous tissue. In all of these characters, it is evident that the Nemertodermatida stands as the plesiomorphic sister group to the Acoela. The peculiar nature of the gut in acoels, therefore, is derived from an epithelial condition.

Studies that have tried to better place the Acoela phylogenetically using techniques of molecular systematics have relied on nucleotide-sequence data in the 18S and 28S rDNA genes. Surprisingly, the 18S data seem to indicate that the acoels are not at all related to the other platyhelminths, not even related to the nemertodermatids. Ruiz-Trillo et al.'s (1999) first study of this kind was sensational enough, the editors of Science thought, to warrant publication in that widely read journal, probably because the authors went so far as to claim that acoels are the most basal of all bilaterians--i.e., that all other bilaterally symmetrical animals (all animals but the sponges and cnidarians) arose from an ancestor that would be classified as an acoel. Such a claim is actually not new; Haszprunar (1996) also proposed a basal position for the Acoelomorpha on the basis of the lack of paired cerebral ganglia, lack of an orthogonal nervous system, and lack of ultrafiltratin nephridia. (With his cladistic analysis, he also splits the remaining platyhelminths into the separate monophyletic groups Rhabditophora and Catenulida and dismisses the phylum as paraphyletic because of its having descendants in the higher spiralians, among others.) Ruiz-Trillo et al. (1999) as well as Littlewood et al. (1999) base their conclusions on 18S-rDNA sequence data. All but one of the 18 acoel species these authors included in this study appeared to have such rapidly evolving 18S rDNA that they could not be used in the analysis that compared the Acoela to other phyla of animals. The only species that appeared not to fall under this restriction was Paratomella rubra, which has, incidentally, been shown by morphological studies (Smith 1981, Smith and Tyler, 1985) to be a phylogenetically important species, so the conclusions about relationships of the Acoela to other flatworms and other phyla rest on it.

This conclusion has been criticized for its lack of consideration of the wealth of morphological characters that contradict it (Tyler et al., 1999). Particularly the notion that the Nemertodermatida is unrelated to the Acoela is denied by many morphological synapomorphies that unite these taxa (Tyler and Rieger, 1977; Smith and Tyler, 1985) as mentioned above; but even the notion that acoels can serve as models for the origin of the Bilateria is contradicted by the highly specialized nature of these worms (Smith and Tyler, 1985). For example, the highly specialized nature of the digestive tissue in acoels is hardly intermediate between the gastrodermis of cnidarians and that of higher animals. So, too, the body wall of acoels,

Amphiscolops gardineri, a relatively large (2-4 mm), colorful acoel

with its lack of extracellular matrix (basement membrane), unique ciliary rootlets, and mesenchymal muscles can hardly be seen as intermediate between that of cnidarians (which have monociliated epitheliomuscular cells resting on a well-developed basement membrane) and such bilaterian phyla as the Annelida and Mollusca. Many of the coelomate bilaterians have epitheliomuscular cells in their coelomic lining, for instance, and if acoels are intermediate then such epitheliomuscular cells would have to have an independent origin in these higher animals (Rieger, 1986). Neither can the complex hermaphroditic reproductive system of the acoels and its involvement in internal fertilization through copulation and a direct life cycle be seen as intermediate between lower diploblasts and the higher animals, many of which, like the cnidarians, have external fertilization through free-spawned gametes and pelagic larvae. The pelago-benthic life cycle, alternating pelagic larvae with benthic adults is widely used by many animals and must be more primitive than the direct life cycle of the acoels (Rieger, 1986, 1994).

The contradiction between the molecular data and the morphological may have arisen because the Acoela is such an ancient group, or because its 18S-rDNA sequence is so rapidly evolving that it has changed to a greater extent than is otherwise the case among animal groups--that is, it is subject in phylogenetic analyses to the so-called long-branch attraction problem (Joffe and Kornakova 2000; Ruiz-Trillo et al. [1999] report that acoels have rates of nucleotide substitution three to five times that of most other animals). Similar results from 18S rDNA have emerged in studies of other phyla, showing, according to Cavalier-Smith (1998), "how grossly misleading the rRNA tree can sometimes be." Jondelius (Jondelius et al., 2002), in collaboration with the same group that reported the acoel 18S rDNA sequences, has reanalysed and applied new sequences from the Nemertodermatida, and they claim to find that this taxon falls in a position as sister group to all bilaterians except the Acoela, which maintains its position as the most basal bilaterian (that is, again, these molecular systematists argue against a taxon Acoelomorpha). Just as Ruiz-Trillo et al.'s (19999) claim about the position of the Acoela is not supported, so, too, is Jondelius et al.'s (2002) claim about the Nemertodermatida not supported. Unfortunately, the rDNA gene is simply not informative about such animals.

The rDNA gene is highly useful, nevertheless, when it comes to deciphering relationships within the Acoela, that is, at lower levels of the taxonomic hierarchy. Hooge et al. (2002) have found, with a larger database of 18S-rDNA sequences from more species of acoels, a branching pattern in the gene tree that is well supported by morphological characters of sperm and body-wall musculature (see pdf of the correlation between the gene tree and morphological characters.

Conclusions

The Platyhelminthes may well prove to be polyphyletic. At the same time, it is likely that all of its taxa originated through a process of progenesis (Rieger, 1986, 1994; Tyler, 2001) even if those are independent origins. No other hypothesis of origins accounts for the histological structure of flatworms. Progenesis has played a major role in much of the evolution of the intersititial fauna (Westheide, 1987), and while this process can be readily recognized among a few taxa in which sufficient intermediates are known (the interstitital annelids, for instance), that for the flatworms seems to have produced such a drastic revamping of morphology that the origins have become obscurred. Because of this morphological obfuscation, the group is a prime candidate for molecular phylogenetics, and one hopes that the appropriate molecules will soon be discovered to decipher its origin(s). Unfortunately, rDNA genes, for which we have so much comparative data, are not among them.

References

Cavalier-Smith T (1998) A revised six-kingdom system of life. Biol Rev 73:203-266

Haszprunar G (1996) Plathelminthes and Plathelminthomorpha -- paraphyletic taxa. J Zool Syst Evol Res 34:41-48

Hooge MD, Haye P, Tyler S, Litvaitis MK, Kornfield I (2002) Molecular systematics of the Acoela (Platyhelminthes) and its concordance with morphology. Molecular Phylogeny and Evolution (in press)

Joffe BI, Kornakova EE (2001) Flatworm phylogeneticist: between molecular hammer and morphological anvil. In: Littlewood T, Bray R, eds. Interrelationships of the Platyhelminthes. Taylor & Francis, London. pp 279-291

Jondelius U, Ruiz-Trillo I, Baguna J, Riutort M (2001) The Nemertodermatida are basal bilaterians not members of Platyhelminthes. Zool Scr 31:201-215

Ehlers U (1985) Das Phylogenetische System der Plathelminthes. Gustav Fischer, Stuttgart, New York. 317 pp

Littlewood DTJ, Rohde K, Clough KA (1999) The interrelationships of all major groups of Platyhelminthes: phylogenetic evidence from morphology and molecules. Biol J Linnean Soc 66: 75-114

Rieger RM, Tyler S, Smith JPS III, Rieger GE (1991) Platyhelminthes: Turbellaria. In: FW Harrison & BJ Bogitsh (eds) Microscopic Anatomy of Invertebrates. Vol. 3: Platyhelminthes and Nemertinea. Wiley-Liss, New York: 7-140

Rieger RM (1994) The biphasic life cycle -- a central theme of metazoan evolution. Am Zool 34:484-491

Rieger RM (1986) Ueber dem Ursprung der Bilateria: die Bedeutung der Ultrastrukturforschung fuer ein neues Verstehen der Metazoenevolution. Verh Dtsch Zool Ges 79:31-50

Ruiz-Trillo I, Riutort M, Littlewood DTJ, Hernieu EA, Baguñá (1999) Acoel flatworms: earliest extant bilaterian metazoans, not members of Platyhelminthes. Science 283:1919-1923

Smith JPS (1981) Fine-structural anatomy of the parenchyma in the Acoela and Nemertodermatida (Turbellaria). Ph.D. Dissertation, Univ North Carolina, Chapel Hill.

Smith JPS, Tyler S (1985) The acoel turbellarians: kingpins of metazoan evolution or a specialized offshoot? In: Conway Morris C, George JD, Gibson R, Platt HM (eds) The origins and relationships of lower invertebrates. Oxford University Press, Oxford, pp 123-142

Smith JPS, Tyler S, Rieger RM (1986) Is the Turbellaria polyphyletic? Hydrobiologia 132:13-21

Tyler S (1999) Systematics of the flatworms--Libbie Hyman's influence on current views of the Platyhelminthes. Am Mus Novitates 3277: 52-66

Tyler S (2001) The early worm--origins and relationships of the lower flatworms. In: Interrelationships of the Platyhelminthes. Littlewood DTJ, Bray R (eds). Taylor & Francis, London. pp 3-12.

Tyler S, Rieger RM (1977) Ultrastructural evidence for the systematic position of the Nemertodermatida (Turbellaria). Acta Zool Fennica 54:193-207

Tyler S, Tyler MS (1997) Origin of the epidermis in parasitic platyhelminths. Int J Parasitol 27: 715-738