GEOPLANIDAE Stimpson, 1857

Sluys, Ronald, Kawakatsu, Masaharu, Riutort, Marta & Baguñà, Jaume, 2009, A new higher classification of planarian flatworms (Platyhelminthes, Tricladida), Journal of Natural History 43 (29 - 30), pp. 1763-1777 : 1773-1774

Family GEOPLANIDAE Stimpson, 1857

Triclads with an auxillary, ventral nerve plexus and a distinct creeping sole.

Since some or all characters that make up the creeping sole may be secondarily lost in some land planarian taxa, this structure requires some additional discussion. The majority of the land planarians possess creeping soles that are presently generally understood to be ciliated creeping soles or creeping ridges, albeit not always specifically referred to as being provided with cilia ( Von Graff 1912 –17; Hyman 1951; Ball and Reynoldson 1981; Sluys 1989b). Generally, there is no mention in the literature of the secretions that are also usually present in a ciliated creeping sole; they are taken for granted. We suggest the following definition of the creeping sole:

A flat or ridged modified strip of epithelium on the ventral surface of geoplanid triclad flatworms characterized by the presence of cilia, the relative predominance of cyanophil glands, and absence of rhabdoids of the rhammite type, and which provides propulsive forces by ciliary or muscular action, or by a combination of both.

Some taxa have secondarily lost the ciliated creeping sole, such as the genera Geobia and Arthurdendyus , together with as yet undescribed Australian taxa (L. Winsor, pers. comm.). The absence of a creeping sole in Geobia was already noted by Von Graff (1912 –17), who postulated that these animals do not practice the creeping or sliding type of locomotion usual in triclads, but wriggle, twist and turn in the manner of nematodes.

Evidently, the new phylogenetic trees imply that the old distinction between Rhynchodeminae and Geoplaninae , based on the presence of only two or multiple eyes is no longer valid. In the new scheme (cf. Alvarez-Presas et al. 2008: fig. 4) the Rhynchodeminae is shown to be the sistergroup of a former geoplanid taxon, the Caenoplanini , with the Rhynchodeminae + Caenoplanini in turn sharing a sistergroup relationship with the Geoplaninae . The caenoplanids studied by Alvarez-Presas et al. (2008) are: three species of Artioposthia , Arthurdendyus triangulatus , three species of Caenoplana , a species of Newzealandia , and a species of Australoplana . From the Geoplaninae they examined four species of Geoplana , and one species of Notogynaphallia . The new classification reflects these recently acquired phylogenetic insights.

In their paper, Carranza et al. (1998: 639) remarked that for the new grouping of the freshwater triclads plus the terrestrial planarians, “perhaps a suitable name... would be the Continenticola .” Although the phylogenetic tree in their paper ( Carranza et al. 1998: fig. 3) plots character states on several branches, they did not present an explicit discussion on the possible diagnostic features of the new clade. Furthermore, the label on the tree refers to “ Continenticola ”. As a consequence, Baguñà and Riutort (2004: table 3) refer to this taxon as “ Continenticola ” nom. nud. (see also Tyler et al. 2006). However, since the International Code of Zoological Nomenclature (ICZN) (1985, 1999) does not regulate taxon names above the family group level a diagnosis is not required in order to make the names available. These names are simply names of groups that may be coined or replaced when deemed necessary. Therefore, we have here assigned the Continenticola to the rank of Subor- der with the authorship of Carranza et al. (1998).

At this stage it is difficult to find unequivocal morphological apomorphies enabling one to provide a diagnosis for the Continenticola . One may be inclined to use the features listed by Sluys as apomorphies for this clade: loss of Haftpapillen (adhesive papillae), presence of resorptive vesicles, reduction of the number of longitudinal nerve cords ( Sluys 1989b: fig. 1, characters 14, 15, 16, respectively). The same features were also presented on the tree of Baguñà et al. (2001: fig. 6.6, characters 8, 9, 10). However, these three presumed apomorphies were originally used by Sluys (1989b) without any consideration of the taxonomic status, phylogenetic position and anatomical features of the Cavernicola , a taxon that was erected one year later ( Sluys 1990). Although the trees presented in Baguñà et al. (2001) and in Alvarez- Presas et al. (2008, fig. 1B) suggest that these three characters may function as apomorphies for the Continenticola , this is only due to the fact that their characters 8, 9, and 10 and 2, 3, and 4, respectively, are positioned at an incorrect level of universality. Loss of adhesive papillae, presence of resorptive vesicles, and reduction of the number of longitudinal nerve cords are features also to be found in members of the Cavernicola . Therefore, these characters should be placed as presumed apomorphies of a clade comprising Cavernicola , the freshwater planarians, and the land planarians (cf. Sluys 1990: fig. 5). As a consequence, it is presently very difficult to find unequivocal autapomorphic characters for the Continenticola . Evidently, the problem hinges on the phyletic position of the Cavernicola as a separate taxon. It is highly opportune to undertake a molecular phylogenetic analysis of all cavernicolan genera in order to determine whether they fall within or outside of the Continenticola . If they fall within the Continenticola the afore-mentioned characters are autapomorphies for the Continenticola (see also Ax 2008), if not, then apomorphies for the Continenticola remain to be discovered.