Csiromedusa medeopolis

Description of holotype. Body considerably flatter than a hemisphere, evenly rounded with a large aboral invagination with a beveled edge. Exumbrella in life covered with faintly visible mosaic of nematocyst patches; when preserved, the mosaic pattern is absent, and instead the exumbrellar surface is sparsely covered with minute warts. Margin of the body scalloped into 8 well defined broad lobes, separated by tentacular peronia. Tentacles 8, arising from the exumbrella in two alternating whorls; one (“ marginal ”) whorl originating from beneath a shallow mesogleal covering, located approximately a quarter of the way up the exumbrella toward the apex; the other (“ apical ”) whorl lacking this shelf, located nearly half the way up toward the apex. Stomach 8 - pointed star-shaped, frosty in appearance, corresponding radially with the 8 tentacles; all stomach pouches reach to point of origin of tentacles. Mouth small, round, simple, approximately one-third the diameter of the stomach; with a slightly thickened margin. Structures that we believe to be gonads are described as follows: vertical in orientation and granular in form, resembling sky-scrapers, icebergs or boulder outcroppings, protruding from the apical conical cavity which is open to the surrounding water, separated from the stomach by a thin membrane; smooth-surfaced in the holotype, consistent with, and therefore interpreted as, testes. Statocysts of the free ectodermal form, 8, interradial. Otoporpae 8, 1 per interradius, located immediately distal to the statocyst, short and narrow. Velum thin and narrow; proximally scalloped into 8 triangular sections, separated by the statocysts; distally continuous. Colouration in life: proximal tips of tentacles and the cells beneath the statocysts crimson; tentacles and stomach translucent whitish; gonad (?) opaque off-white; all other parts transparent and colourless. Variation from holotype. One paratype has only 7 rays, 4 corresponding with the more aboral whorl of tentacles, and 3 with the more oral whorl, while another has 9 rays, with 5 corresponding to the aboral whorl and 4 with the oral whorl. The third paratype has 8 rays, with 4 tentacles to each whorl, and the 4 th paratype is too damaged to count the rays. PLATE 1. Csiromedusa medeopolis gen. et sp. nov., holotype, live, oral view. PLATE 2. Csiromedusa medeopolis gen. et sp. nov., holotype, drawn from life and preserved. A. Oral view. B. Aboral view. C. Lateral view. Scale bar = 0.2 mm. continued next page Bell Diameter Tentacles Manubrium Otoporpae Habitat Source Plankton sample observations. First impression is a small flat whitish disk with a white manubrium approximately half as long as the bell diameter. In fact, the part that looks like the manubrium is what we interpret as the brooding chamber, and it is aboral rather than oral in position. Very inactive, will be found on the bottom of the sorting bowl.

Diagnosis. As for genus.

Remarks. This spectacular medusa is easily distinguished from all other adult and juvenile narcomedusae in having the main tentacles in two whorls, and in having an open apical beveled cavity, containing structures that appear to be gonads. While the peculiar arrangement of these characters could be used to separate the species as a subclass outside the Narcomedusae, it is clearly a narcomedusa in general morphological respects, and we take the conservative approach to recognize it as a narcomedusan pending further insights, particularly those gained from life cycle and molecular analyses, both beyond the scope of this paper. One might wonder whether Csiromedusa is merely the young of another species, based on its small size and fact that some narcomedusae have very strange larval stages. Unfortunately, we do not have a large size (? age) range of specimens that would elucidate the question of maturity. However, the information available on the young of other narcomedusae is convincing that Csiromedusa is unique (see Table 2). The larval stages and development of Cunina and Cunoctantha are surprisingly well studied (McCrady 1859; Metschnikoff 1881; Fewkes 1884; Brooks 1886; Korotneff 1891; Stschelkanowzew 1906; Mayer 1910; Berrill 1950; Bouillon 1987), and Csiromedusa resembles neither the internal medusa-larva stage nor the external parasitic stolon-larva stage. The internal medusa-larva stages of Cunina proboscidea Metschnikoff & Metschnikoff (1871), generally have four tentacles, 20 sensory clubs, no otoporpae, no stomach pouches, and a ring-like gonad in the subumbrellar ectoderm, similar to Solmaris (Mayer, 1910: 478); in contrast, Csiromedusa has two whorls of four tentacles each (total of 8), 8 statocysts, 8 otoporpae, 8 stomach pouches, and apical “ gonads ”. Csiromedusa is more similar to the external parasitic stolon-larva stage of Cunoctantha octonaria (McCrady 1859) and Cunoctantha parasitica Metschnikoff (1881), which have 8 lappets, 8 tentacles (which do not change their position), 8 radial stomach pouches, and 8 marginal sense clubs, with otoporpae developing later (Mayer, 1910: 462); however, Csiromedusa has two whorls of tentacles (unlike the stolonlarvae), conspicuous otoporpae, and the “ gonads ”, whether they are indeed gonads or not, are unusual structures and in a unique location. Furthermore, no trace of the conspicuous protruding manubrium (or ‘ tube buccal’ or ‘ tube manubrial’ of Bouillon, 1987) that immediately distinguishes the parasitic stolon larva is found in Csiromedusa. Berrill (1950) discussed aboral budding in Cunoctantha and Cunina; the descriptions and figures indicate a process completely different to the morphology of Csiromedusa, and thus it does not appear that the apical “ gonads ” of Csiromedusa are medusa buds. However, if they were, this would be truly remarkable and noteworthy. An excellent review of development of the Narcomedusae was given by Bouillon (1987), without any reference to any stage or species with two whorls of tentacles, or with gonads or other structures in a similar arrangement to that of Csiromedusa. The curious feature of the apical opening is currently unclear. Some have suggested to us that it might be from net damage, but the evenly beveled edge would seem to suggest that it is of biogenic origin rather than collecting artifact. It is also conceivable that it is an unclosed umbilical canal; however, we would imagine that if the medusa were still young enough to have a fully open umbilical canal, that it might still have traces of the enormous manubrium characteristic of the stolon larva, which is absent in our specimens. Our specimens have the simple mouth opening that is more characteristic of adult Narcomedusae. Furthermore, the above hypothesis still would not explain what the “ gonad ” structures are or why they are in the cavity. Whatever the origin and function of this cavity, it is nonetheless a unique feature among the Narcomedusae. Similarly, the structures that we have interpreted as “ gonads ” are most unusual. Similar structures and their position are not found in any other adult or juvenile narcomedusa. They are granular and resemble other hydrozoan gonads that we have studied. However, they are in an unexpected location and of an unexpected form. We have made the choice not to sacrifice the few precious type specimens that we have, for histological study, because we believe that the unique two-whorled arrangement of tentacles is sufficient to separate this species from all others. It is our hope that this paper will stimulate further research into examining these unusual structures and answering the questions of their origin and function. Finally, the two-whorled arrangement of the tentacles is of great interest to us. Whilst in some hydromedusae it may be commonplace to develop tentacles in different whorls during ontogeny, this does not appear to be the case for Narcomedusae. Numerous studies of juveniles all describe or illustrate only a single whorl of tentacles (Table 2). Good reviews of narcomedusan development were given by Berrill (1950) and Bouillon (1987). One might also question whether Csiromedusa is more closely related to the Laingiomedusae than the Narcomedusae. Given that the former are characterized by having a conspicuous manubrium with gonads at the base, and no such structures are present in Csiromedusa, this hypothesis would not appear to be supported. While description of this interesting form could be delayed until hundreds of specimens are captured and the life cycle is known, both are more likely to occur if the description is available so that specimens can be recognized when future collections are made. Tasmania boasts a vibrant marine biological research program but gelatinous micro-medusae are not often studied. It is impossible to predict when material sufficient to more fully understand the biology and ecology of this species will be collected, and it would be sad if it were to remain a curiosity hidden on museum shelves.

Distribution. Presently known only from the type locality, i. e., CSIRO wharf, Hobart, Tasmania.

Etymology. The genus name Csiromedusa (pronounced SIGH-row-med-OO-suh) is in recognition of the Commonwealth Scientific and Industrial Research Organization (C. S. I. R. O.), from whose wharf this species was found. The species name, medeopolis (mede- (Gr., genitals), - polis (Gr., a city )), literally, “ city of gonads ”, is in reference to the strange arrangement of the structures that we interpret as gonads, which look like a cluster of tall buildings or icebergs. Noun in apposition.

Material examined. Holotype: TMAG K 3696 (= GZ 0200), CSIRO Wharf, Battery Point, Hobart, Tasmania [42 ˚ 53 ’ 11.5 ” S, 147 ˚ 20 ’ 21.2 ” E], coll. by W. Zeidler and L. Gershwin 1 February 2002; 1.24 mm BD, male. Paratypes: SAM H 1099 (= GZ 0199), same data as holotype, but separate tow; 3 loose specimens, plus 1 specimen mounted on a microscope slide. Additional material: SAM XH 1099 (ex. GZ 0199), same data as holotype, but separate tow; 1 specimen frozen in liquid nitrogen.