[ref. ID; 7064 (James R. Garey et al., 1998)]

Morphological evidence for evolutionary relationships among the Rotifera

Phylum Rotifera consists of three groups, the classes Bdelloidea, Monogononta, and Seisonidea. A three taxon tree has ony three rooted solutions, and each has been proposed at various times for the rotifers.

Tree A is probably the most accepted, because it unites Bdelloidea and Monogononta with a number of characters that are most certainly synapomorphic for the two taxa such as clefts but no pores in the terminal organ of the protonephridia, unpaired retrocerebral glands, salivary glands integrated into the mastax (Ahlrichs 1995, 1997) and the presence of a vitellarium (Wallace & Colburn 1989). In this tree, Seisonidea is the most basal group. Wallace and Colburn (1989) suggested that Bdelloidea + Monogononta be united as the Eurotatoria, and that all three classes make up the phylum Rotiera, while Ahlrichs (1997) only applies the name Rotifera to Bdelloidea + Monogononta.
Tree B has been suggested by Pennak (1989) with Seisonidea and Bdelloidea united as the digonont rotifers (paired female gonads), forming a sister group to Monogononta (unpaired female gonads). Paired gonads are most likely the plesiomorphic condition within Bilateria, and would not unite Seisonidea with Bdelloidea.
Tree C has Bdelloidea as the most basal rotifer with Seisonidea and Monogononta united based on males being present, no bladder, and cellular stomach with microvilli (Ricci et al., 1993). However, these characters are most likely plesiomorphic because they are found in outgroup taxa such as Platyhelminthes. Another character, similarities of the internal layer in the syncytial integument (Clement 1993) has been discussed by Ahlrichs (1997).

Morphological evidence for the evolutionary relationship between Rotifera and Acanthocephala

Although rotifers and acanthocephalans have historically been included among the Aschelminthes (Ruppert & Barnes 1994), it is clear that Aschelminthes is a polyphyletic (Lorenzen 1985; Malakhov 1994; Neuhaus 1994; Winnepenninckx et al. 1995; Ehlers et al. 1996; Wallace et al. 1996) or paraphyletic (Nielsen 1996) assemblage and that the pseudocoelom evolved independently in several aschelminth phyla (Remane 1963; Ruppert 1991; Nielsen 1995). Despite this, a close affinity between Rotifera and Acanthocephala was suspected by Haffner (1950) based on common charaters such as a cloaca, protonephridia, egg segmentation, and muscles that retract the anterior region of the body (Remane 1963). Lorenzen (1985) suggested that rotifers and acanthocephalans can be united based on the internal layer of the syncytial epidermis found in both (Storch & Welsch 1969) and the testis attached to a reduced intestine in monogononts comparable to the ligament cord found in acanthocephalans (Haffner 1950). Lorenzen's analysis did not resolve the relationship between seisonid and monogonont rotifers, but he united Bdelloidea + Acanthocephala based on the presence of lemnisci and similarities of the proboscis in both taxa. These two characters have been rejected by Clement (1993) and Nielsen (1995) as synapomorphies for Bdelloidea + Acanthocephala because the 'proboscis' of acanthocephalans develops from different regions in the embryo than the comparable structure in bdelloid rotifers. The lemnisci are sac-like structures with a high number of lacunes and a still not completely understood function (Miller & Dunagan 1985; Dunagan & Miller 1991), while the structures in bdelloids are most likely thickened regions of the epidermis that carry the rotatory organ (Ahlrichs pers, comm.). However, ultrastructural investigations of this region are still lacking. Nielsen (1995) and Wallace et al., (1996) have both proposed a sister relationship between Rotifera and Acanthocephala, leaving each phylum monophyletic. The character used to group all three classes of rotifers seaprately from acanthocephalans are parthenogenesis, hypodermic impregnation, absence of collagen (Wallace et al. 1996) and toes with adhesive glands (Nielsen 1995). However, many of those characters may not be autapomorphies for Rotifera. Seisonidea reproduce exclusively by sexual reproduction (Clement & Wurdak 1991), so parthenogenesis is not an autapomorphy for Rotifera. Apparently, copulation has never been observed in Seison, which, unlike other rotifers, lacks a penis but has a spermatophore-like structure (Ricci et al. 1993; Ahlrichs 1995). Free sperm cells have been observed only in the reproductive tract of female Seison, and it is likely that sperm enter through the cloaca, so hypodermic impregnation is not likely to be an autoapomorphy for Rotifera. We are not aware of any studies that conclusively demonstrate that collagen is absent from Seison. The presence of toes with adhesive glands as an autoapomorphy of Rotifera has come under question because the cement glands of acanthocephalans may be homologous to the adhesive glands of rotifers (Near et al. 1998). A novel scheme has recently been proposed (Ahlrichs 1997) that most closely relates Acanthocephala with Seisonidea using dense bodies within the spermatozoa and bundles of filaments within the epidermis as synapomorphies. These characters have not before been used for phylogenetic studies and so their significance remains to be confirmed. Ahlrichs retains a monophyletic Rotifera as Bdelloidea + Monogononta, and uses the taxan name Syndermata for Rotifera + Seisonidea + Acanthocephala based on the presence of a syncytial epidermis with an internal layer, outer epidermal cell membrane intrusions with bulbs and a anterior insertion of the flagellum on sperm cells.

Morphological evidence for the position of Rotifera-Acanthocephala within the Bilateria

In most textbooks, rotifers are placed among the aschelminths (e.g. Ruppert & Barnes 1994) or loosely grouped with other pseudocoelomate taxa (e.g. Hyman 1951; Brusca & Brusca 1990). Other views such as a relationship to derived platyhelminth groups (Markevich 1993) are rare. Some cladistic studies of the entier Metazoa consider the pseudocoelom an important character which can result in a monophyletic aschelminth clade (e.g. Schram 1991; Eernisse et al. 1992), but the pseudocoelom has been shown to be a doubtful phylogenetic character (Ruppert 1991). It is clear that the aschelminths are polyphyletic, but the more rigorous treatments of aschelminth taxa often fail to include mainstream metazoan phyla such as arthropods, annelids and molluscs (e.g. Lorenzen 1985; Wallace et al. 1996), although a few recent studies have (Ehlers et al. 1996; Nielsen et al. 1996). Recent morphological analyses from a number of laboratories seem to be converging on the concept of two 'aschelminth' clades, one (Nemathelminthes) containing Priapulida + Kinorhyncha + Loricifera + Nematoda + Nematomorpha + Gastrotricha (Nebelsick 1993; Neuhaus 1994; Ehlers et al. 1996; Nielsen et al. 1996; Wallace et al. 1996), the other clade (Syndermata) containing Acanthocephala + Rotifera (Nielsen et al. 1996; Wallace et al. 1996) and possibly including Gnathostomulida (Ahlrichs 1997). While the Nemathelminthes are most probably the sister group of Spiralia (Lophotrochozoa) within Protostomia (Ehlers et al. 1996), Syndermata + Gnathostomulida (named Gnathifera) have been hypothesized as the sister taxon of Platyhelminthes within Spiralia (Ahlrichs 1995).