June 2004

 

Staurozoa: a new class of Cnidaria?

 

A stauromedusa, Craterolophus convolvulus

At the time that Cubomedusae was separated from Scyphozoa and designated as a class based on distinctive life-cycle features (Werner 1975), members of Stauromedusae were also known to possess an ontogeny (glossary) different from that of the other scyphozoan groups (Uchida 1929). The primary distinguishing traits (glossary) for Cubozoa are metamorphosis of the polyp stage into a single medusa and a lack of polyp features such as a set of four intramesogleal muscles. While the latter features are present in stauromedusans, they lack strobilation (glossary) and instead undergo a single metamorphosis at the oral end of the polyp to produce a benthic, stalked adult. Cubomedusae became a class and Stauromedusae did not because of a subjective argument. The former had characteristics deemed to be sufficiently different to suggest that they did not belong in Scyphozoa.

Today, systematic decisions such as separating one group from a containing taxon are based upon phylogenetic criteria rather than on the distinctiveness of features. Two recent phylogenetic analyses, one a cladistic analysis of morphology and life-history characters (Marques and Collins 2004) and the other an analysis of 18S rDNA (Collins 2002), indicate that Stauromedusae does not form a clade with the other scyphozoan groups, Coronatae, Rhizostomeae, and Semaeostomeae. In both analyses the latter three groups form a clade for which strobilation and ephyrae are likely synapomorphies (glossary) .

Based upon these results, Marques and Collins (2004) suggest two systematic actions that are rather different in nature. The first suggestion was to limit the name Scyphozoa to be a reference to the clade characterized by strobilation, in following with the idea that taxonomic names should only be applied to monophyletic entities. The second systematic decision was to erect a fifth class within Cnidaria called Staurozoa, consisting of Stauromedusae and Conulatae, an extinct taxon (Cambrian to Triassic) with pyramidal skeletons. But invoking a Linnean rank might seem strange during an age when systematics is guided by phylogenetic criteria.

While it is widely appreciated that Linnean ranks are arbitrary, they may convey important information in the context of a classification based on phylogenetic hypotheses. Obviously, comparison of classes of cnidarians with classes of chordates is difficult and perhaps nonsensical because no objective standard for what constitutes a class exists. On the other hand, recognizing Staurozoa as a class within Cnidaria communicates the hypothesis that Staurozoa is a primary clade of cnidarians that is historically discrete, and correspondingly distinct in its biology, from Anthozoa, Cubozoa, Hydrozoa and Scyphozoa. In other words, recent phylogenetic findings suggest that Cnidaria is more diverse at a fundamental level than previously appreciated. Recognizing Staurozoa as a new class communicates this finding in a simple and relatively straightforward manner.

 

The oral surface of Craterolophus convolvulus

What is distinct about the biology of Stauromedusae? Well, for starters, the ultrastructure of stauromedusan ovaries and ocelli are profoundly different from that seen in coronates, rhizostomes, and semaeostomes (Eckelbarger et al. 1993, Blumer et al. 1995). In fact, stauromedusan ovaries are more complex than in any other cnidarian group. The life cycles of stauromedusans are also markedly peculiar. Rather than possessing free-swimming medusae, adult stauromedusans live attached to the substrate by a stalk. The planulae of stauromedusans are unlike those of other cnidarians in being non-ciliated and with an invariant number of stacked, non-dividing endodermal cells (Otto 1976, 1978). These cells are contractile and move the planula along in a creeping inchworm-like fashion. After creeping for a number of days, the larva settles and develops into a juvenile polyp. Later, the oral end of the polyp metamorphoses and takes on a number of characters that resemble those in adult medusae of cubozoans and scyphozoans, including hollow structures of tentacular origin (rhopalioids), circular coronal muscles, gonads, and ocelli (Kikinger & Salvini-Plawen 1995). At the same time, the aboral portion of the adult retains polypoid characters such as gastric septa and longitudinal muscles.

Might these latter characters be synapomorphies for a clade containing Stauromedusae, Cubomedusae, Coronatae, Rhizostomeae, and Semaeostomeae? This is a distinct possibility and the optimal tree obtained by Marques and Collins (2004) would suggest that this is the case. Prior to Cubozoa becoming a class, these five taxa constituted Scyphozoa and therefore this name, rather than being constrained to refer to the latter three groups (as suggested by Marques and Collins), could have been altered to once again have its more traditional meaning. However, published molecular results (Collins 2002) do not support such a clade, though they do not strongly contradict its existence. Molecular data from new sources (both taxa and markers) are being gathered in order to clarify the precise positions of Stauromedusae, Cubomedusae, and the clade containing the strobilating scyphozoans.

Anatomy of Stauromedusae (from Mayer 1910)