Bzohgical Journal of the Linnean Society (1993). 48: 239-266. With 5 figures Zoogeography and life cycle patterns of Mediterranean hydromedusae (Cnidaria) Dipartimento di Biologia, Stazione de Biologia Manna, Universita di Lecce, 73100 Lecce, Italy AND J. BOUILLON Laboraioire de <oologie, Universitk Libre de Bruxelles, Ave F.D. Roosevelt 50, 1050 Bruxelles, Belgique Received July 1990, accepted far publication December 1991 The distribution of the 346 hydromedusan species hitherto recorded from the Mediterranean is considered, dividing the species into zoogeographical groups. The consequences for dispersal due to possession or lack of a medusa stage in the life cycle are discussed, and related to actual known distributions. There is contradictory evidence for an influence of life cycle patterns on species distribution. The Mediterranean hydromedusan fauna is composed of 19.5% endemic species. Their origin is debatable. The majority of the remaining Mediterranean species is present in the Atlantic, with various world distributions, and could have entered the Mediterranean from Gibraltar after the Mcssinian crisis. Only 8.0% of the fauna is classified as Indo-Pacific, the species being mainly restricted to the eastern basin, some of which have presumably migrated from the Red Sea via the Suez Canal, being then classifiable as Lessepsian migrants. The importance of historical and climatic factors in determining the composition of the Mediterranean fauna of hydromedusae is discussed. ADDITIONAL KEY WORDS:-Hydrozoa - hydroid - medusa - dispersal CONTENTS Introduction . . . . . . . . . . . . . . . . The Mediterranean Sea . . . . . . . . . . . . Zoogeographical regions . . . . . . . . . . . . Biological features of hydromedusae affecting their distribution . . . Types of hydromedusan life cycle and their possible relevance to dispersal The Mediterranean hydromedusan fauna . . . . . . . . Material and methods. . . . . . . . . . . . . . Results . . . . . . . . . . . . . . . . . Circumtropical species . . . . . . . . . . . . Endemic species . . . . . . . . . . . . . . Boreal species . . . . . . . . . . . . . . Cosmopolitan species . . . . . . . . . . . . . Tropical-Atlantic species . . . . . . . . . . . . Mediterranean-Atlantic species . . . . . . . . . . Indo-Pacific species . . . . . . . . . . . . . 0024-^066/93/030239 +28 $08.00/0 239 Q 1993 The Linnean Society of London 240 F BOER0 AND J. BOUILLON . . . . . . . . . . . . . . Discussion Affinities of the Mediterranean hydromedusan fauna . . . . . . . . The importance of life-cycle features in the distribution of Mediterranean hydromedusae . . . . . . . . . . . . . Conclusions . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . References . . . . . . . . . . . . . . . Appendix . . . . . . . . . . . . . . . 252 252 253 255 256 256 258 INTRODUCTION The description and explanation of the distribution of organisms is the main goal of biogeography. T h e theoretical framework on which biogeographical analyses are usually based can be divided into two approaches: the historical and the ecological. The historical approach implies that the distribution o i a species reflects its evolutionary history, so that biogeographic and phylogenetic analyses are strictly related by taking speciation processes into account. The original version of the historical approach consists of the dispersal theory, typically accepted by Darwinian and Neo-Darwinian evolutionary biologists. According to this theory species originate mainly by allopatry due to dispersal of members of the ancestral population into separate geographic areas. The actual distribution of organisms is thus the result of the dispersal of their ancestors, being due to the intrinsic potential of the species, in terms of vagility. In recent years the theories of historical biogeography have been changed by a different approach which has led to vicariance biogeography, with the integration of Croizat's panbiogeography (e.g. Croizat, 1964) and the cladistic method (e.g. Nelson & Platnick, 1981 ) . Vicariance biogeography postulates that ancestral species were widely distributed before the fragmentation of the Pangaea and that continental drift led to physical separation of the original populations, leading then to speciation. The actual distribution of organisms, in this case, is due to extrinsic reasons, being caused by the fragmentation of the areas inhabited by their ancestors. A further development of this second approach led to cladistic biogeography (e.g. Humphries & Parenti, 1989), with the construction of cladograms for both phylogenies and areas of distribution. A contradiction between these two theories is, however, apparent. They take into account two aspects of the history of life which are not mutually exclusive. The fragmentation of the Pangaea undoubtedly had a decisive impact on both speciation and distribution of organisms, but the possession of a wide distribution throughout the Pangaea ( a starting point in vicariance biogeography) implied dispersal from a centre of origin or, less probably, instantaneous speciation on a worldwide scale. The ecological approach (see, for instance, Davis, 1982; Endler, 1982; Blondel, 1986) states that historical factors are not necessarily linked to actual distributions, and that species are present in those localities where their ecological requirements are satisfied. Of course this condition must be valid for every theory, and also in this case there is not a real conflict with the historical approaches. Vermeij (1978) attempted to reconcile historical and ecological approaches in marine zoogeography. These problems, however, have been tackled mainly in terrestrial organisms. Marine zoogeography, even in the light of the most modern approaches, is still MEDITERRANEAN HYDROMEDUSAE 24 1 largely devoted to the determination of actual distributions. The unexplored portions of the world ocean are so vast that the descriptive stage is far from being completed (van der Spoel, 1983). Fishes and some invertebrates such as molluscs have better known distributions, but this is usually not the case for the rest of the faunas. Especially in invertebrates, marine biologists have mainly stressed the evolution of life-history traits and their relevance to dispersal (e.g. Valentine & Jablonski, 1983; Strathmann, 1985; Jackson, 1986), or have based the explanation of species distributions in terms of adaptations to local conditions e.g. Vermeij, 1978, 1989). For some groups, such as molluscs, however, lifehistory traits have been used to explain distributions (e.g. Scheltema, 1986) and evolution (e.g. Jablonski, 1986). Hydromedusae, in both their hydroid and medusa stages, occur commonly in all oceans and seas but a synthesis of their world distribution has never been attempted (see Kramp, 1959, 1961, 1968 for the medusa stage only). They have much-differentiated cycles (see below), and almost all reproductive and dispersal strategies of higher animals are already shown at the hydromedusan evolutionary level. We have chosen to study the relationships of the hydromedusan fauna of the Mediterranean Sea because it is one of the better known in the world and because the geological history of the basin has been recently carefully reconstructed. The opening of the Suez Canal, connecting the Mediterranean and the Red Sea, furthermore, constitutes a rare opportunity for 'experimental' biogeography. After a general description of the history and the physical conditions of the Mediterranean, and of the life-cycle types of hydromedusae, we will consider the affinities of the Mediterranean hydromedusan fauna, trying then to compare the actual distributions with the results expected by the application of the different biogeographic theories. The Mediterranean Sea The Mediterranean Sea is considered to be a relict of the Tethys Sea, the body of water separating Gondwana and Laurasia following the fragmentation of Pangaea. It connected the early Atlantic and Pacific Oceans. During the Miocene (Pontian) the eastern part of the Tethys Sea closed, and the only communication left was that with the Atlantic Ocean. When this connection closed as well, the Messinian crisis (between 6 and 5 Ma BP) led to the almost complete drying of the Mediterranean. Only the deeper parts of the basin seem to have retained water (see Maldonado, 1985 for a review of the geological history of the Mediterranean). Salinity, and probably temperature, were very high. The opening of the Strait of Gibraltar (5 Ma BP) restored the level of the sea. The Mediterranean relicts of the Tethys Sea, therefore, would have passed the Messinian crisis in almost non-marine conditions or in refuge areas. This possibly led to many local extinctions of both flora and fauna. The sea-grass Posidonia is the most outstanding case of Tethyan endemism: representatives of this genus live only in the Mediterranean and in Southern Australia. How the ancestral stock of the single Mediterranean species, Posidonia oceanica, survived the Messinian crisis is still debated and the same questions apply to the remainder of presumed Tethyan species (see Perks, 1985, for a discussion). 242 F. BOERO AND J. BOUILLON Recent studies (see Por, 1989, for a review) are showing that the Messinian crisis was perhaps not so drastic throughout the basin as previously thought, so this topic is to be considered as not completely clarified. Today the Mediterranean communicates with the Atlantic via the Strait of Gibraltar and with the Red Sea via the Suez Canal, opened in 1869. The physico-chemical conditions of the Mediterranean are different from those of the Atlantic Ocean and the Red Sea. Deep-water temperature is constant at about 13'C. This is the mean temperature of the whole basin in the cooler part of the year (January-March), with slightly higher values in the eastern basin and very low values (4-5%) in the northern Adriatic. Surface temperature can reach 28OC in August. In shallow waters, then, the temperature differences between the warm and the cold season can approach 15-20°C. Salinity is about 37%0, and so is higher than in the Atlantic (about 35Oh) and lower than in the Red Sea (40-41%). The eastern basin has salinities of up to 39%. Strong seasonality is thus a striking feature of the Mediterranean. Temperature is the most variable factor, accompanied by variations in a number of other physical factors, including the concentration of nutrients, water movement and light penetration. A 'warm' season (May-June to October-November) thus alternates with a 'cold' season (November December to AprilMay). Planktonic and benthic primary and secondary production show sharp seasonal cycles reflecting this alternation of climatic conditions. ~ ~ g e o g r a p h i c aregions i Marine zoogeography is fairly advanced in some groups (especially vertebrates) but lower invertebrates such as Hydrozoa have received scant attention. The incompleteness of our knowledge even of the overall distribution of hydromedusae is exemplified by the situation in the Pacific insular area. The synopsis by Kramp (1968) is the standard work for the area and lists 59 species of Antho- and Leptomedusae. A long period of observation at a single site in Papua New Guinea raised the number to 176, with the description of 43 new species and 96 new records from the area (Bouillon, Clareboudt & Seghers, 1986). Some of these newly described species are now being found in the Mediterranean! It is hence inadvisable to divide the oceans into detailed regions and subregions. The distribution patterns considered (Mediterranean Endemic, Mediterranean-Atlantic, Boreal, Tropical-Atlantic, Indo-Pacific, Circumtropical, Cosmopolitan) are, for convenience, taken to have the Mediterranean as their centre and are compared with it (Fig. 1). They apply then to the Mediterranean fauna and consider all the possible relations between this and other faunas. For instance, we consider as Indo-Pacific the species found both in the Mediterranean and the Indo-Pacific, even though an Indo-Pacific species should not necessarily occur in the Mediterranean. Biological features of hydromedusae affecting their distribution Hydromedusae are represented by a medusa, a planula and a polyp stage. The alternation of benthic and pelagic stages is a basic feature of hydromedusae, though in some orders the polyp is absent (some Narcomedusae and all MEDITERRANEAN HYDROMEDUSAE .--. *" - *. Figure 1 . Zoogeographical regions for the Mediterranean hydromedusan fauna: A, Mediterranean-Endemic; B, Mediterranea-Atlantic; C, Boreal; D, Tropical-Atlantic; E, Indo-Pacific; F, Circumtropical; Cosmopolitan not shown. (Redrawn after C. N. Bianchi, unpublished.) Trachymedusae). Almost half of the Mediterranean species, however, have lost the medusa stage by reduction (Table l ) , so that a much varied array of dispersal strategies is present in this group. In this paper we consider the planktonic medusa as the sexual, adult stage: it releases the gametes, giving rise to non-feeding planula larvae from which, in most cases, originate hydroids (a specialized type of larva) which, then, will produce medusae (see Boero & Bouillon, 1987; Boero & Sara, 1987; and Bouillon, Boero & Fraschetti, 1991, for recent discussions). This interpretation, however, is not accepted by other hydromedusan workers (Cornelius, 1990). When the medusa is present in the cycle, the adult shows the highest vagility and could be considered as the main agent of dispersal. This is a reversal of the 'norm' in meroplanktonic animals, where the larva, and not the adult, has a planktonic life. F. BOER0 AND J. BOUILLON 244 TABLE I. Distribution of Mediterranean hydromedusae 7- Yo " % g % mg Yo Endemic Mediterranean Atlantic Boreal Tropical Atlantic Indo-Pacific Circumtropical Cosmopolitan Non-classifiable Total T,Total number of species referable to a given contingent and %, percentage ofthat contingent in respect to the total fauna; m, number of species with medusa and %, percentage of such species within the contingent; g, number of species with fixed gonophores and %, percentage of such species within the contingent; mg, number of species with liberable eumedusoids and O/', percentage of such species within the contingent. *, Significant difference ( x 2 test, P < 0.05) between species with medusae and species with fixed gonophores or liberable eumedusoids; same difference, but highly significant (P < 0.01). + Types of hydromedusan life cycle and their possible relevance to dispersal (Fig. 2 ) ( 1 ) Medusa - planula - benthic hydroid - medusa Dispersal is dependent on several factors: duration of life of the medusa (from a few days to several months), duration of the free life of the usually hollow and floating coeloblastula and planula larva (too few data for generalization, but a maximum of 15 days seems to be possible), dispersal of hydroid via planktonic propagules (e.g. Halecium pusillurn), life on nectonic (Hydrichthys), or planktonic (Obelia, Kinetocodium, Pelagiana, parasitic Narcomedusae etc.) organisms, and settling on floating algae or objects. A particular case is that recently reported by Bouillon et al. ( 1991 ) in Laodicea indica, a leptomedusa producing planulae which, according to the season, can give rise to hydroid colonies or short-lived fixed gonothecae which immediately release a single medusa. ( 2 ) Liberated eumedusoid or swimming gonophore - planula - benthic hydroid - liberated emedusoid or swimming gonophore The planktonic life of liberated eumedusoids or swimming gonophores (reevolved medusiform stages) is usually just a few hours (see Boero & Bouillon, 1989) limiting the extent of dispersal. As in the former cases, however, the hydroid can contribute to dispersal. (3) Benthic hydroid - planula - benthic hydroid The coeloblastula is absent and the morula and planula stages of these species are usually dense and tend to sink. The possibilities for dispersal are thus limited. In a few species the planula is known to be linked to the mother colony by mucous threads which break only when settlement occurs. I n some groups a non-feeding actinula larva occurs, showing some possibilities of dispersal. Production of asexual propagules, life on pelagic organisms or on floating objects, are still possible. MEDUSA POLYP encystment possible -to +++ feedina complete -to ++ + 1 feedina reduced + abolished DISPERSAL non-feedina re-introduced + n.f. n. f. ++ PLANULA . encystment possible -to +*+ MEDUSA 1 non -feeding Figure 2. Life-cycle patterns of hydromedusae, with dispersal possibilities (from- 10 + 1, presence ( + ) or absence f - ) of asexual reproduction, and trophic value of the various stages Broken arrows: direi-t development, with no hydroid stage; solid arrows' indirect development, with hydroid stage. 246 F. BOER0 AND J. BOUILLON (4) Medusa - planula - planktonic hydrozd - medusa The benthic life is abolished and different dispersal strategies are employed by planula, hydroid and medusa stages (e.g. Margelopsis, Pelagohydra, Velella, Porpita, Climacocodon, Evens hexanemalis). (5) Medusa - planula - medusa This is considered a primitive type of life cycle and it is characteristic of most of the Narcomedusae and all Trachymedusae. Besides exceptional benthic forms (Ptychogastria), all species are holoplanktonic. Asexual reproduction of medusae The life span of medusae should set a limit on their dispersive capabilities. But this is compensated by several ways of asexual reproduction such as fission and budding of medusae from the manubrium or tentacular bulbs, gonothecae on the circular or radial canals, polyps on the manubrium or radial canals. In this way a medusa and its offspring should be able to cover unlimited distances, provided that food availability and chemico-physical features of the water are suitable (Kramp, 1959; Bouillon et al., 1986; Mills, 1987). Encystment Almost all hydroids are able to produce resting stages represented by dormant hydrorhizae (Calder, 1990). Several species are known to produce planula encystment and this phenomenon is probably more widespread than is known. Recently Carre & Carre (1990) have described the asexual formation of resting frustules from the medusa of Eucheilota paradoxica. Specimens capable of such encystments can survive for long periods and become active again under proper conditions. When the possibility of hydroids settling on floating objects (including ships) is considered, it is evident that, theoretically, dispersal has no limit (Cornelius, 1981;Jackson, 1986). These life-cycle patterns should generate different dispersal possibilities, so that it might be possible to classify them along an efficiency-of-dispersal gradient. Picard (1958) and Boero (1984), however, have remarked that lifecycle features seem unimportant in determining the distribution patterns of hydromedusae. The most efficient cycle for dispersal we could envisage apriori is one with both medusa and benthic hydroid. The two completely different dispersal and feeding strategies, plus the planula stage, enable a wide array of possibilities, even though not all species presumably can express the maximum theoretical dispersal potential. The cycle of Laodicea indica, with the possibility of shifting from benthic hydroid to benthic gonotheca, can be placed in this category. It might be expected that species with such a life cycle would show a low rate of endemism, with a high tendency to wide distribution. The second position might be held by species having free medusae and planktonic hydroids. They cannot take advantage of settlement and encystment on floating objects, but are anyway able to disperse with two morphs having different dispersal and feeding strategies. Holoplanktonic species, with the medusa stage only, rely on a single morph which, however, has no limitation due to the finding of a proper substratum for MEDITERRANEAN HYDROMEDUSAE 247 larval settlement. The possibility of resting stages is only hypothetical and their distribution is limited by food availability and physico-chemical conditions. The lowest vagility is shown by species with liberable eumedusoids, swimming gonophores and, above all, fixed gonophores. Their dispersal is mainly due to the planula displacement but since their larval stages are solid and usually nonfloating, the covered distances should not be relevant. The dispersal of propagules and resting stages deriving from the hydroid, however, is still theoretically rather high. We are aware that this scenario is oversimplified. The general biology of the great majority of the species is still unknown and, furthermore, the life cycles of about 75% of the species are still to be elucidated. It is to be expected that species with no medusa stages show a greater tendency to dispersal by asexual propagules or simply by colony rafting, but it is also true that species with medusae can show planula settlement on substrata such as pteropods, fishes etc., so that there should be a certain balance among the different dispersal mechanisms. We will try to test the preceding assumptions against the known distribution of the Mediterranean hydromedusae, assigning them to zoogeographical groups and considering their life cycles. The analysis will be hindered by incomplete knowledge of distributions and also by the fact that some areas have been extensively investigated for medusae but not for hydroids, and vice versa. It has been impractical to build up a group for each category of life cycle, and we choose to divide the species into forms with medusae, forms with fixed gonophores, and forms with liberable eumedusoids or swimming gonophores. Species with pelagic hydroids and Trachymedusae (with no hydroid stage) constitute a small fraction of the whole fauna: for ease of analysis they have been considered as species with both hydroid and medusa stages. Until now all species of hydromedusae are supposed to have a polyp stage, with the exception only of some Narcomedusae and the Trachymedusae. But the life cycles of 82 of the 143 Antho- and Leptomedusae species with medusae are unknown or poorly known. As suggested by Bouillon et ul. (1991) it could be that many or at least some species known only as medusae have no 'classical' polyp stage. Thus our speculations are based on incomplete knowledge, but it is also true that the study of Mediterranean hydromedusae has been, and still is, rather intense and that the Mediterranean is one of the best known hydromedusan faunas of the world. The number of species treated here probably constitutes a sufficiently large sample to allow some general considerations. The knowledge of the distributions of many of them will surely improve, but this will take place slowly and this is not a sufficient reason to delay delineation of general aspects of species distribution. The Mediterranean hydromedusanfauna By hydromedusae we mean practically all Hydrozoa except Siphonophorae, that is: Antho-, Lepto-, Laingio-, Limno-, Narco- and Trachymedusae, and the Actinulidae (see Bouillon, 1985, for definition of orders). No recent paper, to our knowledge, has treated the complete hydromedusan fauna of the Mediterranean. Kramp (1959, 1961) treated the medusa stage only F. BOER0 AND J. BOUILLON 248 . (65 species), and Ficard (1958) considered both polyp and medusa stages of Antho- and Leptomedusae (191 species). The preparation of a monograph on Mediterranean hydromedusae has also contributed to the knowledge of the group. T h e only part to have been published is that on the Anthomedusae/Capitata (Brinckmann-Voss, 1970). Many recent papers have greatly modified the knowledge of the composition of the hydromedusan fauna of the Mediterranean, with new records and descriptions of new families, genera, and species. These, at first, were considered endemic to the basin, but many have since been recorded from other seas and oceans. MATERIAL AND METHODS The distribution of the representatives of the various orders is summarized in Tables 1 and 2. With the y2 test we tested the significance of the difference in numbers between species with medusae and species with fixed gonophores, swimming gonophores and liberable eumedusoids. We included the species with swimming gonophores or liberable eumedusoids in the group of species with fixed gonophores for a number of reasons: (1) the possibility could be high that there are more species of Leptomedusae liberating TABLE 2. Distribution o f the different orders of Mediterranean hydromedusae. An~homedusae Endemic Mediterranean Atlantic Boreal Tropical Atlantic ndo-Pacific Circumliopical Cosmopolitan Non-classifiable 'total Leptomedusae Endemic Mediterranean Atlantic Boreal Tropical Atlantic Indo-Pacific Circumtropical Cosmopolitan Non-classifiable Total Anthomedusac~Leptomedusae Endemic Mediterranean Atlantic Boreal Tropical Atlantic Indo-Pacific Gircumtropical Cosmopolitan Xon-classifiable Total MEDITERRANEAN HYDROMEDUSAE 249 Limnomedusae Endemic Mediterranean Atlantic Boreal Tropical Atlantic Indo-Pacific Circumtropical Cosmopolitan Nan-classifiable Total Narcomedusae (all m) Endemic .Vedi terranean Atlantic Boreal Tropical Atlantic Indo-Pacific Circumtropical Cosmopolitan Non-classifiable Total Trachymedusae (all m) Endemic Mediterranean Atlantic Boreal Tropical Atlantic Indo-Pacific Circumtropical Cosmopolitan Non-classifiable Total Actinulidae (all g) Cosmopolitan Mediterranean Atlantic Total Laingiomedusae (all m) Indo-Pacific Total 1 1 100 0.2 Abbreviations as in Table I. gonophores than hitherto thought (Boero & Bouillon, 1989); (2) some species may switch from fixed gonophores to liberable eumedusoids, according to environmental conditions, so the two possibilities are not mutually exclusive (Stefani, 1959); (3) the possibilities of dispersal obtained with a liberable eumedusoid are presumably more similar to those obtained with fixed gonophores than to those obtained with medusae. RESULTS The number ofhydromedusan species recorded from the Mediterranean is 346 (see Appendix). The number of species with medusae is not statistically different F. BOER0 AND J. BOUILLON Distribution Figure 3. Species numbers of Mediterranean hydromcdusae. E, Endemic; MA, Mediterranean-Atlantic; B, Boreal; TA Tropical-Atlantic: IP, Indo-Pacific, GI Circumtropical; C, Cosmopolitan; NC, non-classifiable. T , Total, M, species with medusa; G, species with fixed gonophorca; MG, species with liberable eumedusoids or swimming gonophores; *, Significant difference (x2 test, P < 0.05) between species with medusae and species with fixed or swimming gonophores or liberable eumedusoids; **same difference, but highly significant [P < 0.01). than that without medusae. The situation, however, is completely different when the most abundant groups are considered separately. Anthomedusae show a significant prevalence of species with medusae, whereas in Leptomedusae the species with fixed gonophores or short-lived medusoids prevail (Table 2). The different zoogeographical groups (Figs 3-5) .are treated separately in order of importance. Czrcumtropical specie's T h e circumtropical species are the most abundant, with a highly significant prevalence of species with medusae. The Anthomedusae show a highly significant difference in favour of the medusa stage; the Leptomedusae show no significant difference between species with and without medusae; all LimnoNarco- and Trachymedusae have a well-developed medusa stage. The data show that the medusa stage is dominant in the circumtropical Anthomedusae, but not in the Leptomedusae. D Q C 3 T M G M G Distribution Figure 4. Species numbers of Mediterranean Anthomedusae. Key as in Fig. 3. MEDITERRANEAN HYDROMEDUSAE Distribution Figure 5. Species numbers o f .Mediterranean Leptornedusae. Key as in Fig. 3. Endemic species T h e endemic contingent is almost as important as the circumtropical one. There is no significant difference between species with and without medusae in Antho- and Leptomedusae. All Narco- and Trachymedusae have medusae. T h e presence of a medusa stage in the life cycle of endemic Mediterranean hydromedusae is rather widespread. Boreal species T h e overall difference between species with and without medusae is not statistically significant. Leptomedusae, however, show a statistically highly significant difference in favour of fixed gonophores. As in the circumtropical contingent, Antho- and Leptomedusae show differing life-cycle patterns, Anthomedusae being 'neutral', whereas Leptomedusae show a relevant reduction of the medusa stage. I t is suggestive that the orders with prevalence of the medusa stage (Narco-, Trachymedusae) have no boreal representatives in the Mediterranean. Cosmopolitan species We reluctantly introduce this category which should comprise panoceanic species occurring from the Polar seas to the Equator, I t is unlikely that such species really exist, and their records in the literature could be due to insufficient possibilities of discrimination in current taxonomy. Many of the supposed cosmopolitan species may turn out to be eurythermic circumtropical, but this sort of zoogeographical revision is outside the scope of the present paper. T h e difference between cosmopolitan species with and without medusae is statistically highly significant in favour of fixed gonophores. Cosmopolitan Anthomedusae, however, show no significant difference whereas Leptomedusae show a highly significant difl'erence for fixed gonophores; all Narco-, Limno- and Trachymedusae have medusae whereas the Actinulidae have fixed gonophores. The Mediterranean species with the broadest distributions show a sharp 252 F. BOERO AND J. BOUILLON prevalence of forms deprived of the medusa stage and, again, Antho- and Leptomedusae behave in an opposite way. Tropical-Atlantic species There are no significant differences between species with and without medusae. This coiltingent does not show a significant prevalence of a given type of life cycle. Mediterranean-Atlantic species There is no overall significant difference between species with and without medusae. However, the Anthomedusae have a statistically significant difference in favour of the medusa stage and the Leptomedusae have a statistically highly significant difference in favour of fixed gonophores; Limno-, Narco- and Trachymedusae, all have medusae; the only representative of the Actinulidar has fixed gonophores. The presence of the medusa stage is different in Anthoand Leptomedusae, the two orders showing opposite life-cycle patterns. As in some of the preceding cases, this is compensated in the overall picture, so that the presence or absence of the medusa seems unimportant. hdo-Pacific species This group of species shows a highly significant difference in favour of the medusa stage. The difference, however, is not significant for Leptomcdusae. The presence of a medusa stage in the life cycles is widespread in the Indo-Pacific species inhabiting the Mediterranean, but not in the Lcptomedusae, in which the situation is balanced. DISCUSSION Affinities o f the Mediterranean hydromedwan j'auna The endemic group is second only to the circumtropical one. This indicates a great originality of the Mediterranean fauna. As remarked by Picard (1958), however, the only certain endemics are those species restricted to particular habitats not available outside the Mediterranean. Posidonia oceanica meadows constitute an outstanding example (Boero, 1987). Many endemic species have been found only once, in spite of intense collection in the basin in recent decades. Their endemicity could be due to incomplete knowledge of their distribution. These species could have arrived in the Mediterranean from other, less studied, areas where they are more abundant but still undetected. Some of the endemic species seem to be restricted to the Adriatic which, in fact, is a quite peculiar sea. Its conditions might have facilitated speciation. The environmental conditions of the Mediterranean, as already mentioned, are very variable during the year and this should favour forms with a marked tendency towards seasonality, such as hydromedusae. Warm-water species can proliferate in the summer and pass the winter as resting stages. Cold-water species could be active in the winter and spend the summer as resting stages. This pattern is evident from studies of hydromedusae, of both hydroid and MEDITERRANEAN HYDROMEDUSAE 253 medusa stages (see Boero, 1984, for a review, and Morri & Bianchi, 1983, for a discussion of brackish water species). Some of the endemic species could be relicts of the Tethys Sea. This can apply to the species typical of Posidonia since this plant is supposed to be itself a Tethyan relict. Paraco~yneh e i could be a Tethyan relict too, and features of its life cycle (Bouillon, 1975) could have enabled it to survive the Messinian crisis. Not many other species are easily classifiable in their endemicity. As already said, they could be 'false endemics', due to sparse zoogeographical information, but they could also have originated in the Mediterranean after, or during, the Messinian crisis (see Pkrks, 1985; Sara, 1985; Tortonese, 1985; Por, 1989, for recent discussions). Some endemic species are of dubious taxonomic validity, owing to insufficient description. The difference in salinity between the Mediterranean and the Atlantic could play a role in the confinement of stenohaline species which evolved in the Mediterranean Sea. Dispersal of specimens settled on floating objects or of strictly shallow-water species could be influenced by the fact that, owing to the differences in density, the Atlantic water enters the Mediterranean basin from the surface, whereas the Mediterranean water flows out at a deeper level. Differences in salinity and features of circulation could be the main causes for the confinement of the species which evolved in the Mediterranean. One hundred and twenty-six species are boreal, tropical Atlantic, or Mediterranean-Atlantic; and 114 species are circumtropical or cosmopolitan. Almost 70% of the hydromedusan fauna living in the Mediterranean could have entered through the Strait of Gibraltar, having been found i n t h e corresponding part of the Atlantic and also elsewhere. Indo-Pacific species are noteworthy, representing only 8.0% of the fauna. Picard (1958) stated that no Indo-Pacific species was present in the Mediterranean, but the studies of Schmidt (1973, 1976), Marinopulos (19791, Lakkis & Zeidane ( 1985), Goy, Lakkis & Zeidane ( 1 9901, Margulis (1989) and others have shown that certain Indo-Pacific species are present in the Mediterranean, mainly in the eastern part. This may be due to Lessepsian migration through the Suez Canal, even though the absence of information about the hydromedusan fauna of the Eastern Mediterranean before the opening of this waterway allows no comparison between the situation before and after the presence of a connection between the Mediterranean and the Red Sea. The hydromedusan fauna of the Mediterranean, then, comprises a conspicuous Atlantic contingent which, presumably, is the result of colonization through the Strait of Gibraltar. A relatively high number of endemics gives originality to the fauna, but it is difficult to ascertain their geographical origin, even though some species could be Tethyan relicts. Lessepsian migration via the Suez canal is slowly bringing Indo-Pacific species into the basin and it is expected that this group will become increasingly reported in the near future, following better exploration of the Eastern Basin. For a detailed treatment of Lessepsian migration see Por (1989). The importance of life-cycle features in the distribution of Mediterranean hydromedusae The hypotheses resulting from our analysis of life-cycle features are only partly confirmed by our data. Circumtropical species show a prevalence of cycles with a medusa, but cosmopolitan species behave in exactly the opposite way and fixed 254 F. BOER0 AND J. BOUILLON gonophores prevail over medusae. T h e endemic species should have shown a sharp tendency towards medusa suppression. This is true for the species living on Posidonia leaves, but the whole endemic hydrornedusan fauna shows no significant difference between the two general types of life cycle. T h e data regarding the single orders are even more contradictory. T h e opposite patterns of dominance of species with and without medusae indicate that Anthomedusae show a sharp tendency to conserve the medusa stage, whereas most Leptomedusae have suppressed it. This could be explained by some differences in colony organization between thecate (leptomedusan) hydroids and athecate (anthomedusan) hydroids. Thecates often have highly integrated colonies, formed by a high number of small polyps, whereas athecates usually have bigger polyps and less integrated colonies. A sharp specialization of the hydroid stage could have led to its prevalence over the medusa stage in Leptomcdusae. This assumption, however, is probably too simplistic and the problem calls for a deeper analysis which is outside the scope of the present paper. T h e presence of the medusa stage, and so high degree of vagility, seems not to be of importance in the patterns of distribution of the Mediterranean hydromedusae as already observed by Picard (1 958). Cornelius (1981), analysing the distribution of boreal hydroids, found that two-thirds lacked a medusa stage, being, however, widely distributed in cooler parts of the northern hemisphere. The tendency to medusa reduction is evident also in the present data on Mediterranean species of boreal affinity, whereas the species in other zoogeographical groups do not show this feature. Furthermore, even though not explicitly stated, it is apparent that Cornelius (1981) dealt mainly with records of the hydroid stage, and this may limit the general value of his assumptions. Asking the question if a medusa stage is 'better' for dispersal than fixed gonophores is comparable to asking if planktotrophic larvae are more efficient for dispersal than lecithotrophic ones. Following a series of mathematical expressions Vance (1 973: 35 1 ) summarized his results with this sentence: 'Planktotrophy is more efficient than lecithotrophy when planktonic food is abundant and planktonic predation is low, and lecithotrophy is more efficient when either or both of these conditions is reversed'. I t is quite possible that during daily, lunar, seasonal, annual and pluriannual cycles environmental conditions might be successively better and worse for the different types of dispersal mechanisms available to the various species. Over geological time this should result in a uniform geographical distribution of nearly all marine species, regardless of their dispersal mechanisms. This has obviously not occurred (see van der Spoel, 1983, for discussion). In our opinion the distribution of marine species or, a t least, of hydromedusae does not depend on their modes of dispersal, but on their limits of environmental tolerance. I t is possible that, in one of the many different ways listed above, hydromedusae can widely disperse in the various oceans and seas. The absence of a given species from a certain area may not depend on its not reaching it, but on its lack of adaptation to local conditions. Over short periods, however, the presence of a long-lived pelagic stage seems to be a successful mode of dispersal, as indicated by the predominance of species with medusae in the Indo-Pacific contigent that, presumably, is the result of a recent migration of species from the Red Sea to the Mediterranean. MEDITERRANEAN HYDROMEDUSAE 255 CONCLUSIONS Historical factors have undoubtedly been important in recruitment to the Mediterranean hydromedusan fauna. Species which entered the basin from the Strait of Gibraltar after the Messiniari crisis largely determined the present day fauna, together with a set of palaeoendemisms of Tethyan origin. The peculiar conditions of the Mediterranean, then, led to speciation and neoendemism. This interpretation is in accordance with the one detected in the Mediterranean benthos by Fredj (1974) and in the Mediterranean plankton by Furnestin (19.79). This overall picture seems to reconcile quite well with the dispersal theory, but it is notable that the possession of theoretically more or less efEcient means of dispersal seems not to be important in the determination of the distribution of the species. Recent migration from the Red Sea through the Suez Canal, however, shows that efficient dispersal has a great importance in short-term colonization of newly-available areas. Among Indo-Pacific forms, representing probable Lessepsian migrants, in fact, species with medusae are significantly more numerous than species with fixed gonophores, even though this is not true for Leptomedusae. Por (1981) proposed a Lessepsian Province in the Eastern part of the Mediterranean in direct contact with the Suez Canal, this Province being characterized by a high number of Indo-Pacific species which had migrated to the Mediterranean via the Suez Canal. The migration occurred in spite of temperature and salinity barriers. Dispersal can clearly play a major role in determining the distribution of marine species. The same can be said for environmental features. Lessepsian migrants colonized the Mediterranean because they were able to reach it and because they are adapted to live in a 'Mediterranean' environment. Efficiency of dispersal is important during the first stages of colonization (prevalence of species with medusa stage in the group which entered via Suez) but seems unimportant over geological time ('balanced' situation in the species which entered via Gibraltar). The theory of vicariance could possibly apply to the endemic species living on the leaves of Posidonia, but a comparison of the hydroids of Mediterranean and Australian Posidonia has still to be done. The two theories explaining biogeographical patterns can both be applied to subsets of the hydromedusan fauna of the Mediterranean. Climatic factors, however, play an important role in 'shaping' a given fauna. Recent advances in biogeography (vicariance and cladistic biogeographies) refer almost entirely to terrestrial florae and faunae. It is reasonable to assume that oceans and mountains are almost insurmountable barriers for many terrestrial forms, but the situation in the seas is completely different, and geographical barriers are probably much less important in determining speciation and distribution patterns of marine organisms. Fauchald (1984) rightly stated that, theoretically, any organism can reach any point in the world ocean, in spite of its 'history'. For these reasons we consider premature, for instance, the comparison of the wellknown hydromedusan fauna of the western Mediterranean with that of the eastern Mediterranean or with that of deep waters of the basin: their data sets are simply not comparable. The same is true for comparisons of the hydromedusan fauna of the Mediterranean with those of the Red and the Black seas. As remarked by Sara (1985) the understanding of the causes of the 256 F. BOER0 AND J. BOUILLON distributions of marine animals will be possible by taking into account not only historical aspects (theories of vicariance and of dispersal) but also the conditioning of the present-day environmental features. The statement by Strong (1983: 640); 'Until autecological facets of existence are understood, it is tenuous to infer much about synecological influences' is, in our opinion, applicable also to marine zoogeography in terms of distribution of single species vs composition of regional faunas. ACKNOWLEDGEMENTS This paper was written with contributions from UNEP, MURST (60% and 40% programs), the Fonds de la Recherche Fondamentale Collective nr. 2.9008.90, C. N. Bianchi (La Spezia), D. Calder (Toronto), P. F. S. Cornelius (London), K. Mangin (Tucson), C . Morri (Genova), M. Sari (Genova) and H. Zibrowius (Marseille) read and commented on the manuscript. C. N. Bianchi (La Spczia) proposed the broad zoogeographical regions here adopted. REFERENCES Blondel J. 1986. Biogeographic evalutiue. Paris: Masson S. A. Boero F. 1984. The ecology of marine hydroids and the effects of environmental factors: a review. Pubbtkaz.iouz delta Stazione +ologica di N a p l i I: Marine Ecology 5; 93-1 18. Boero F, 1987. Evolutionary implications of habitat selection in the hydroids ofPosidonia oceanica meadows. In: Bouillon J, Boero F, Cicogna F, Cornelius PFS, eds. Modem trends in the systematics, ecology, and evolution of hydroids and hydromedusae. Oxford: Clarendon Press, 251-256. Boero F, Bouillon J. 1987. Inconsistent evolution and paedomorphosis among the hydroids and medusae of the Athecatac/Aiithomedusae and the ThecataejLeptomedusae (Cnidaria, Hydrozoa). In: Bouillon J, Boero F, Cicogna F, Cornelius PFS, eds. Modem trends in the systematics, ecology, and evolution of hydroids and hydromtdusae. Oxford: Clarendon Press, 229-250. Boero F, Bouillon J. 1989. An evolutionary interpretation of anomalous medusoid stages in the life cycle of some Leptomcdusae (Cnidaria). In: Ryland JS, Tyler PA, eds. Reproduction, Genetics and Distributions of Marine Organisms. Fredenshorogh: Olsen & Olsen, 37-41. Boero F, Sara M.1987. Motile sexual stages and evolution of Leptomedusae (Cnidaria). Belletlino di Zwilsgia 54: 131-139. Bouillon J. 1975. Sur la reproduction et l'kologie de Paraco~nehwd ('I'uhularoidea, Athecataj. Archives de Biologie 86: 45-96. Bouillon J. 1985. Essai de classification des hydroides-hydromeduses (Hydrozoa-Cnidaria). Indo-Malayan zoology 2; 2&-243. Bouillon J, Boero F, Fraschetti S. 1991. The life cycle of Laodicea indica Browne, 1905 (Laodiceidae, Hydromedusae, Cnidaria). In: Williams RB, Cornelius PFS, Hughes RG, Robson EA, eds. Coelenterate 6tology.' recent research en Cnidaria and Ctenophora. Hydrobifsiogta 2161217: 151-157. Bouillon J, Qareboudt M, Seghers G. 1986. Hydromeduses de la Baie de Hansa (Mer de Bismarck; Papouasie Nouvelle-Guinke). Ripartition, conditions climatiques et hydrologiqucs. Indu-Malqn zoology 3: 105-152. Bouillon J, Boero F, Cicogna F, G4li JM,Hughes RG. 1992. Nun siphonophoran Hydrozoa: what are we talking about? In: Bouillon J, Boero F, Cicogna F, Gili JM, Hughes RG, eds. Aspects of hydrozoan bzology. Scienlia Marina 56(2): 279-284. Brmckmann-Voss A. 1970. Anthomedusae/Athecatae (Hydrozoa, Cnidaria) of the Mediterranean. Part 1: Capitata. Fauna e Flora del Golfs di ~Vapoli39; 1-96. Calder D. 1990. Seasonal cycles of activity and inactivity in some hydroids from Virginia and South Carolina. Canadian Journal of ^palogy W: 442450. Carre D, Card C 1990. Complex reproductive cycle in Eucheilota paradoxica (Hydrozoa, Leptomedusae): medusae, polyps and frmtules produced from medusa stage. Marine Biology lIMb 303-310. Cornelius PFS. 1981. Life cycle, dispersal and distribution among the Hydroida. Porcupine Mewsietter 2È 47-50. Cornelius ITS. 1990. Evolution in leptolid life-cycles (Cnidaria: Hydroida). Journal of ~VaturaiHistory 24: 579-594. Croizat L. 1964. Space, time, form: a biolqical synthesis. Caracas: published by the author. MEDITERRANEAN HYDROMEDUSAE 257 Davis GM.1982. Historical and ecological factors in the evolution, adaptative radiation, and biogeography of freshwater mollusks. Amen'can zoologist 22: 375-395. Endler JA. 1982- Problems in distinguishing historical from ecological factors in biogeography. American zoologist 22: 44 1-452. Fauchsfcid K. 1984. Polychaete distribution patterns, or: can animals with Palaeozoic cousins show large-scale geographical patterns? In: Hutchings PA, ed. Proceedings of the First International Polychuete Conference, Sydney: The Linnean Society of New South Wales: 1-6. Fredj G. 1974. Stockage et exploitation des donnkes en hcologie marine. C- Considkrations biogkographiques sur Ie peuplement benthique de la Mediterranhe. M k i r e s de PImtitut ocianographique, Monaco 7: 1-88. Furnestin ML. 1979. Aspects of the zoogeography of the Mediterranean plankton. In: van der SpoeI S, Pierrot-Bults AC, eds. Zoogeography and diversity of plankton. Utrecht: Bunge Scientific Publishers, 191253. C o y J, Lakkis S, Zeidane R. 1 M . Lm mkduses de la Mkditerranie orientale. Bulletin de 1'Insfitut ~cianographique,Monaco, numero special 7: 79-88. H1xxxxphries CJ, Parent! LR. 1989. Cladistic biogiography. Oxford: Clarendon Press. Jablonski D, 1986. Larval ecology and macroevolution in marine invertebrates. Bulletin of Marine Science 39s 565-587. Jackson JBC. 1986. Modes of dispersal of clonal benthic invertebrates: consequences for species' distributions and genetic structure of local populations. Bulletin of Marine Science 39:588-606. Kramp PL. 1959. The Hydromedusae of the Atlantic Ocean and adjacent waters. D a m Reports 46: 1-283. K r a m p PL. 1961. Synopsis of the medusae of the world. Journal @/theMarine Biological Association @/theUnited Kingdom 40: 7-469. Kramp PL. 1968. The Hydromedusae of the Pacific and Indian Oceans (1). Dana Reports 63: 1-200. Lakkis S, Zcidane R. 1985. Les hydromeduses des eaux neritiques libanaises: composition et distribution. Rapports st Procek Verbaux de la Commission international pour l'exploration scienfifique de la Mer Miditerranhe 29; 179-180. Maldonado A. 1985. Evolution of the Mediterranean Basins and a detailed reconstruction of the Cenozoic Paleoceanography. In: Margalef R, ed. Western Mediterranean. Oxford: Pergamon Press, 17-59. Marplis P. 1989. New hydroid jellyfishes of the family 'I'ubulariidae (Coelenterata, Hydrozoa). <oologichtskii zhurnal 68: 126-130. [In Russian). Marinopulos J. 1979. Biological survey of the eastern Mediterranean sea: hydroids (preliminary study). Rapports et Precis Verbaux de la Commission intentalionale pour Isexploration scientijque de4a Mer Mkditerranle 25/26; 119-120. Mills C. 1987. In-situ and shipboard studies of living hydromedusae and hydroids: preliminary observations of life-cycle adaptations to the open ocean. In: Bouillon J, Boem F, Cicogna F, Cornelius PFS, eds. Modern trends in the systematics, ecology, and evoluUon of hydroids and hydro~rtedusae.Oxford: Clarendon Press, 197 -207. Morri C, Bianchi CN. 1983. Contribute alla conoscenza degli idrozoi lagunari italiani: idropolipi del delta del Po (nord Adriaticoj. Aid del Muses civzco di Storia naturale di Trieste 35: 185-205. Nelson GJ, Platnick N1. 1981. Systematics and biogeography: cladistics and vicariance. New York: Columbia University Press. P6rGs JHM. 1985. History of the Mediterranean biota and the colonization of the depths. In Margalef R,ed. Western Mediterranean. Oxford: Pergamon Press, 198-232. Picard J. 1958. Origioes et affinitks de la faune d'hydropolypes (Gymnoblastes et Calyptoblastes) et d'hydrom~duses (Anthomt?duses et Leptomcduses) de la Miditerranhe. Commission internationale pour i'exploratwn scktifique de la Mer Mfditewaafe, R a p w t s et Prods Verbaux des R h i u n s , 14 (ffouv. S k . ) : 187-199. Par FD. 1981. The Lessepsian biogeographic province of the Eastern Mediterranean. Jeurnies d'Etudes w t m a d q u e s et Biogiographiques sur la Mfditarranie. Monaco: Commission internationale pour l'exploration scientifique de la Mer Mhditerranhe, 81-83. Por, FD. 1989. 7-hlegacy el fithys. An aquatic biogecgraphy of the &van&. Dordrecht: Klnwer Academic Publishers. S a d M. 1985. Ecological factors and their biogeographic consequences in the Mediterranean ecosystems. In: Moraituu-Apostolopoulou M, Kiortsis V, eds. Mediterranean marine ecosystems. New York: Plenum Press, 1-17. Scheltema RS. 1986. Long-distance dispersal by planktonic larvae of shoal-water benthic invertebrates among central Pacific islands. Bulletin a/ Marine Science 39s 241-246. Schmidt H. 1973. Hydromedusae from the eastern Mediterranean sea. Israel Journal of ?ool~ey 22; 151-167. Schmidt H. 1976. A comparison of the Hydroida fauna of the eastern Mediterranean and the Red Sea. Ada Adriafka 18: 259-266. Spoel, van der S. 1983. Patterns in plankton distribution and the relation to speciation: the dawn of pelagic biogeography. In: Sims RW, Price JH, WhaIIey RES, eds. Euolutiq time and space: the emergence o f fhe biosphere. London: Academic Press, 291-334. Stefani R. 1959. Sulla variahilita ecologica di un Idrozoo (Campanularia caliculata Hincks), Bollettino di zooiogia - 26: 115-120. Strathmaxm I¬. 1985. Feeding and non feeding larval development and life-history evolotion in marine invertebrates. Annual Renew of Ecology and Systematics 16; 339-361. Strong D. 1983. Natural variability and the manifold mechanisms of ecological communities. American Naturalist 122: 56-80. 258 F. B O E R 0 AND J. BOUILLON Tortonese E. 1985. Distribution and ecology of endemic elements in the Mediterranean fauna (fishes and cchinoder~tis).In: Moraitou-Apostolopolou M, Kiortsis \\ eds. Mediterranean marine ecosystems. New York: Plenum Press. 57-83. Valentine JW, Jablonski D. 1983. Speciation in the shallow sea: general patterns and biogeographic controls. In: Sims RW,Price H, Whalley PES, eds. Evolution, time ami space: the emergence of the biosfiftere. London: Academic Press, 20 1-226. Vance RR. 1973. O n reproductive strategies in marine benthic invertebrates. American Naturalist 107; 339-352. Vermeij GJ. 1978. Biogeography and adaptation. Patterns of marine lift. Cambridge: Harvard University Press. Vermeij GJ. 1989. Inieroo-anic differences in adaptation: effcrts ol'history and productivity. Marine Ecology &ogres:, Series 57: 293-305. APPENDIX List of hydromedusan species hitherto recorded from the Mediterranean Sea. g: fixed gonophores; ni: medusae; mg: liberable medusoids or swimming gonophores; ?: life cycle unknown or poorly known; E: endemic; MA: Mediterranean Atlantic; B: boreal; TA: Tropical Atlantic; IP: 1ndo-Pacific; CT: circun~tropical;C: cosmopolitan; nc: non classifiable. species marked with an asterisk (*) were added to the list wlien the present paper was in press and have not been considered in the text. T h e adopted classification reflects suggestions recently proposed in the paper by Bouillon et al. (1992). Class HYDROZOA Subclass HYDROIDOMEDL'SAE Order ANTHOMEDCSAE 142: 85 m, 52 g, 5 mg Suborder Filifera Superfamily Bougaiiivillioidea Bougainvilliidae 20: 11 m, 9 g Bimeria wstita Wright, 1859 (*}Boue(iinviilia aurantiaciz Bouillon; 1980 Boueainvillia britannica (Forbes, 1841) K o ~ i n i ~ z l l imaniculata a Haeckel, 1864 Beuginmiilia multirilia (Haeckel, 1879'; Boqaimillia ma-fcus {Allman, 1863) Bougainvillia piapgaster (Haeckel. 1879;) Claiwfiselia michaeli (Berrill, 1948) Dico~yneconferta (Alder, 1857) D i c o p e rorybcari (Allman, 1864) Garseia franciscam (Torrey, 1902) Garzieia g~i.ieaMotz-Kossowska, 1905 Carveia nutarts (Wright, 1859) KoetliAerina fasciculata (Pkron & Lesneur, 1810) Liz& blondina Forbes, 1848 Lizziafulgurwts (A. Agassiz, 1865) Lzzzia octosIvIa (Haeckel, 1879) Nubiella mitra Bouillon, 1980 Pachycordyle nafiolitana Weismann, 1883 Rhizorhagium arenosum (Aldcr, 1862) Thamnostoma dibaiiwn (Biisch, 1851 ) Cytaeidae 6: 4 m, 2 g Q l a & fmszlla Gcgcnbaur, 1857 Cytaeis tetrastyla Eschscholtz, 1829 Cytaeis vulgaris Agassiz & Maycr, 1899 Parafvtaeis odonia Bouillon, 1981 Perarelia pop(igt;lala Bavestrello, 1987 Perarella schneideri (Motz-Kossowska, 1905) E CT IP IP E E Superfamily Clavoidae Clavidae 7: 2 m, 5 g C1az:a muUicornis (Forskal, 1775) Car&o#hora caslfia (Pallas, 177 1) Cor~ylo#horapusilla Motz-Kossowska, 1905 m? m'? m? m? g g . MEDITERRANEAN HYDROMEDUSAE Cofydendrium parasiiicum ( L., 1 767 ) Mrrona cornucopiat (Norman, 1864) Oceania armaiw Koclliker, 1853 Turri~ospisnutriiula McCrady. 1859 Superfamily Eudendrioidea Eudendriidae 1 3 13 g Eudmdrium m a t u r n Tichomiroll', I887 Eudendriian ralceolatum Motz-Kossowska, 1905 Eudmdrium capi/lare Alder, 1856 Eudtdrium carneum Clarkc, 1882 Eudefidrkm c u n n k h a m i Kirkpatrick, 1910 Eud&ium Jragile Motz-Kossowska, 1905 EudendtI~tn;glumeratura Picard, 1952 Kudmdrium merulum Watson, 1985 Euder^riurrz ma~&ossowskae Picard, 1952 Eudendrzum rucwt~sum(Gnlelin, 1791) Euderidrium rameum (Pallas, 1766; Er~dtndnumramosum (L.,1758) *ma amboinense Pictet, 1893 Superfamily Hydractinodea Hydractiniidae 14: 8 m, 3 g, 3 mg eydractiuia aaileata (Wagner, 1833) Hydractinia echwala Fleming, 1828 &drachma j'ucicoiu (M. Sam, 1857) Hydrwiinia ornata Bonnevie, 1899 Hfdructomma prmoti (Motz-Kossowska, 1905) hdacoyne areola& (Alder, 1862) Psdworyir hurealis (Mayer, 1900) Podecoyne carnea M. Sars, 1846 Podawrym exigua (Haeckel, 1879) Podocinyne hartlaubi Neppi & Stiasny, 191 1 Podwmyne minima (Trinci, 1903) Podocar~inemimla (Mayqr, 1900) ,Y@fdaria inrrmis (Allman, 1872) Tregoubovia atentaculda Picard, 1958 Ptilocodiidae I: 1 g 77if.cocodium brzeni Bouillon, 1967 Rhysiidae I: 1 g in' Rhysiu halecii (Hicksun & Gravely, 1907) Stylastcridae 1: 1 g Errina ospcfa (L., 1767) Superfamily Pandeoidea Niobiidae 1: 1 m ffiobia dendratentaculata Mayer, 1900 Calycopsidae 2: 2 m Bythotiara muwayz Gunther, 1903 Calycapsis simplex Kramp & Damns, 1925 Pandeidae 13: 13 rn Amphintma dinema (Peron & Lesueur, 1810) Amphimma rubrum (Kramp, 1957) A m p h i m rtigosum (Mayer, 19(10) h j f h i n e m a ium;la (Mayer, 1900) Leuckariiara nohilis Hartlaub, 1913 Leuckurilara octona (Fleming, 1823) Merga guileri Brinckmann, 1962 Merga tergoslina (Neppi & Stiasny, 1912) Merga tregoubovi Picard, 1960 Merga violacea (Agassiz & Mayer, 1899) .Veotwr& pileala (Forskal, 1775) Octatiura russelli Kramp, 1953 Pandta conica (Quoy & Gaimard, 1827) F. BOERO AND J. BOUILLON 260 Protiaridae 2: 2 m Halitiara formosa Fewkes, 1882 (*) Haliliara inzexa Bouillon, 1980 Protiara kiranana (Pefon & Lesueur, 1810) Trichydridae 1: 1 m Trichydra oligonema (Kramp, 1955) Superfamiiy Rathkeuidea Rathkeidae 1; 1 m Rathkea octopnnctata (M. Sara, 1835) Suborder Capitata Superfamily Acauloidea Superfamily Corynoidea Cladonematidae 1: I m Corynidac 18: 12 m, 6 g C q n e caespe.~Allman, 187 1 C o p e epi^oica Stechow, 1921 Cnrynefucicola De Filippi, 1866 C'eryne musmtdcs (L., 1761) C o p e pusilia Gaertner, 1774 C o p e pintmri Schneider, 1897 Dicodonium adriaticurn Graeffe, 1884 Dicodunium ocellatum (Busch, 1851) Dipureaa dolichogaster (Hacckel, 1864) Dipurena haiterafa (Forbes, 1846) Dtpurma ophiogaster (Haeckel, 1879-80) D i p u r m reesi Vannucci, 1956 ¥Sarsi.a eximia (Allman, 1859) Sarsia gemmifera Forbes, 1848 Sarsia producta (Wright, 1858) Sarsia prol@ra Forbes, 1848 Sarsia tubutosa (M.Sars, 1835) Eleutheriidae 3: 3 m Eleatheria clapartdei Hartlaub, 1889 Etmlheria dichotorna Quatrelages, 1842 Staurdadia porlmanni Brinckmann, 1964 Superfamily Moerisioidca Moerisiidae 5: 5 m Moerisia carinae Bouillon, 1981 MoenSia inkermanka Paltschikowa-Ostroumova,1925 Moerisia lyonsi Boulenger, 1908 Moerisia paltasi (Derzhavin, 1912) Odessia maeotica (Ostroumoff, 1896) Protohydridae 1: 1 g Protohydra feuckarti Greef, 1869 * ) Spaeroeorynidae ) Spaerocorym bedoti' Picket 1893 Superfamily Tricyclusoidea Fricyclusidae 1: I g Tricyclusa singularis (Schulze, 1876) Superfamily Tubularioidea Boreohydridae 1: 1 g Psammafydra nana Schultz, 1950 MEDITERRANEAN HYDROMEDUSAE Corymorphidae 9: 8 m, 1 g Branchiocerianthw italicus Stechow, 1923 Coiymorpha nutarns M. Sam, 1835 Eucodoniam brownei Hartlauh, 1907 EupA,sora annulala Kramp, 1928 Euphysora bigelowi Maas, 1905 @ybvcodon prolifer L. Agassiz, 1862 Paragolhea bathybia Kramp, 1942 Plotocnide borealis Wagner, 1885 Vannucctaforbesii (Mayer, 1894) Euphysidae 2: 1 m, 1 g Euphysa auratu Forbes, 1848 (*) Euphysaftammea (Linko, 1905) SifihonoA,dra adriatica Salvini-Plawen, 1966 Halocordylidae 1: 1 mg Halocordyle tlisttcha (Goldfuss, 1820) Paracorynidae 1: I g Paracnryne huvei Picard, 1957 Tubulariidae 9: 6 m, 3 g Ectspleura dumcrtieri (Van Beneden, 1844) Eclopleura minerzu Mayer, 1900 Ectopleura sacculifera Kramp, 1957 Ectopleura wrighti Peterscn, 1979 Eugothoea petalina Margulis, 1989 Rhabdomi singularis Keferstein & Ehlers, 1861 Tubularia crocea Agassiz, 1862 Tibuiaria indivisa L., 1758 Tubutaria larynx Ellis & Solander, 1786 %. Superfamily Porpitoidea Porpitidae 2: 2 m Porpiia prplla (L., 1 758) Velella vefella (L., 1 758) Superfamily Zancleoidea Cladworynidae 1: 1 g Cladoco~ynffloccosa Rotch, 187 1 Halocoryne epizoica Hadzi, 1917 Rosalindidae 1: I m Rosalinda incruslans (Kramp, 1947) Zancleidae 2: 2 m <anclea coslaia Gegenbaur, 1857 m e a sessilis (Gosse, 1853) Order LEPTOMEDUSAE 154: 57 m, 93 g, 4 mg Suborder Conica Infraorder Campanulinida Superfamily Campanulinoidea Aeq uomidae 4: 4 m Aequorea forskalea PCron & Lesueur, 1810 Aequorea conica Browne, 1905 Aequorca pensilis (Eschscholtz, 1829) Zygocama sp. Babnik, 1948 Blackfordiidac I: 1 m Biackfoda ukginica Mayer, 1910 262 F. B O E R 0 AND J. BOUILLON Campanulinidae 4: 4 g Cdycella M g a (I.., 1767) Cam~~anulina hincksii Har tlaub, 1897 Egmundella amiranlen.~~.~ Millard & Bouillon, 1973 Lafoeina f e w & G.O. Sam, 1874 h&tldgazziidae 1: 1 m Oi-tophtaluciumfunfranum (Quay & Gaimard, 1827) Phialellidae 1: 1 n1 Phialella qnadrafa (Forbes, 1848) Superfamily Dipicurosomtoidea Melicertidae 1: I m Orfhislumella graejfei (Neppi & Stiasny , 1911) Superfamily Eirenoidea Eirenidae 9: 8 m 1 mg Eirtne viridula (Peron & Lesueur, 1810) Eusymnanlhea kquilina inquilimi Palombi, 1935 Euiimo pegenbasri (Haeckel, 1864) Eutima &lis {Forbes & Goodsir, 1853) Eutima mira McCrady, 1859 Eutonina sciniillarvs (Bigelow, 1909) Helgicirrha c a n (Hacckcl, 1864) Helgicirrho schulzei Hartlaub, 1909 i't'eotima lucullana [Delle Chiajc, 1822) Laudireidae 8: 8 m Kramptlia dubia Russell, 1957 Laodicea higeiortii Neppi & Stiasny, 19 12 Laodiceajijiana Agassiz & \layer. 1899 Uodicea oceliata Babnik, 1948 I s o d i ~ e aneptuna Mayer, 1900 Laodicea unduiata (Forbes & Goodsir, 1851') Melicerlisw adnatica Neppi, 1915 Staurophora mertensii Brand t, 1838 Tiarannidae 1: 1 m Mooderza rotunda (Quoy & Gaimard, 1827; riaropsiidae 2: 2 rn Ocl^gonade mediterranea Zoja, 1896 Tiaropsulium mediterranenm (Metschnikoff, 18861 Superfamily Lovenelloidea Cirrholoveniidac I: 1 m Cirrholovenia tetrunema Krarnp, 1959 Eucheilotidae 3: 3 m Eucheilota maasi Neppi & Stiasny, 1911 Ewheiloia paradiixica Mayer, 1900 Eucheilota zcniricularis MrGrady. 1859 l^iMnella chiquitita Millard, 1959 Losenella cirrata (Hacckel, 1879) iJiuenella ciausa (Loven, 1836) Uvaulla pacilis (Clarke, 1882) Lmmtlla fmniculata 5 .0. Sars, 1873) Superfamily Mitrocomoidea Mitrocomidae 2: 2 m A¥fitrocoma annae Haeckel, 1864 Milrocomella howini (Kramp, 1930) MKDHERRANEAN HYDROMEDUSAE Infraorder 1.afoeida Superfamily Lafoeidea Hcbcllidae 7: 4 m, 2 g, 1 mg Hebella hrochi {Hadzi, 1913) Hebdla JUrax %Hard, 1957 Hebella parasilica ('Ciamician, 1880) Hebella scadens (Bale, 1888) Hebdla urrmiata Millard, I964 Arandia gigas ( Pieper, 1884) Srandia michael-sarsz {Lcloup, 1935) .4cyptolaria conftrta ( Allman, 1877) FtUllum serpew (Hamall, 1848) Filellum serratum (Clarke, 1879) Laf'a~adumosa (Fleming, 1820) Lafvea fruti~osa (Sars, 18.51) Zy^ephylax hiarmak Billard, 1905 111f~orcicr Haleciida Superfamily Halecioidt'a Haleciidac 18: 1 m, 16 g, 1 mg Campaleciurn medusiferum Torrey, 1902 Hdecium banyuleme Moiz-Kossowska. 1911 Heleciwn beanii ijohnston, 1838) Halecium coriicum Stcchow, 1919 Hueleriurn halecinm (L., 1758) HaSerium {airosum Alder. 1859 H e l d lankesteri (Bourne, 1890) Ildecium medihrraneum Weisinann, 1883 Halecium mnricaium [Ellis & Solander, 1786) Halrcinm nunum Alder, 1859 Halecium peirosum Stechow, 1919 Hafecium pusillurn (M. Sars, 1857) Halfcimn sessile Norman, 1867 Halecium tenellam Hincks, 1861 Hydranthea aliysii (Zoja, 1893) Hydranthea margarica (Hincks, 1863; Ophiodissa cacinijurmis (Ritchie, 1907' Ophiodissa mirahilh (Hincks, 1868) Infraorder Plumulariida Superfamily Plumularoidea Aglaophcniidae 15: 15 g Aglmph~niaacacia Allmian, 1883 Maojheniu elongala Meneghini, 1845 Aglaaphenia harpago Von Schenek, 1963 .4g~aoflAeniakkhenpauer: (Heller, 1868) Aglaophenia latecarinala Allman, 1877 Aflaophenia lophacarpa Allman, 1877 Aglaophenia picardi Svoboda, 1979 Aglaophmia pluma (L., 1758) Aglaophnia octodonta (Heller, 1868j Aglaophenia tuhijomis (Marktanner-Turncretscher, 1890) .4~~laopACTiu tubulifira (Hinds, 1861) Cladocarpus dallfusi BiIlard, 1924 Thecocarpus distans (Allman, 1877) Thecocarpus mynopAyllum ( I..., 1758) Thecocarpus phyteuma (Kirchcnpauer, 1876) Halopteriidae 7: 7 g Antendla secufida& (Gmelhi, 1791) Anteniulla siliquosa (Hincks, 1877 ) Halopleris caiharina (Johnston, 1833) IP E c E c B TA MA B TA MA TA c c E B CT MA m? g g g K g g B K g g g g g g me, R g 264 F. BOER0 AND J. BOUILLON Hafopteris diaphana (Heller, 1868) Haloptern giutinnsa (Lamouroux, 1816) Halopteris litchtensterni (Marktanner-Turnerets~her,1890) S~hi~otricha frutescens (Ellis & Solander, 1786) Kirchenpaueriidae 3: 3 g Kkchenpaueria echinulata (=ricks, 1868) Kirchmpawria pinnata (L., 1758) Ventromma haleciaides (Alder, 1859) Plurnulariidae 7: 7 g h e r t e s i a aniennina (L., 1758) Nernertesia ramosa Lamouroux, 1816 ffemertesia tctrasticha (Meneghini, 1845) Plumularia obliqua (Thompson, 1844) Plumularia putcheffa Bale, 1882 Plumuluria setacea (L., 1758) Plumularia syriaca Billard, 1930 Superfamily Sertularioidea Sertulariidae 22: 22 g Amphisbeha optrculata (L.,1758) Diphasia margarcta (Hassall, 1841) D p m e n a disticha (Bosc, 1802) Salacia desmoides (Torrey, 1902) Salacia dubia (Billard, 1922) Sertufarella arbuscula (Lamouroux, 181 6 j Sertularella crassicaulis (Heller, 1868; Sertularella cubica Garcia, Aguirre & Gonzalez, 1980 Strtularella cyldritheca (Allman, 1888) Sertularclla fusiformis (Hincks, 1861) Strtularella gaudichaudi f Lamouroux, 1824) Sertularella gayi fLarnouroux, 1821) Strtularella picta (Meyen, 1834) Sertularella potyzonias (L., 1758) Sertularella robust0 Coughtrey, 1876 Sertularilla simplex (Hutton, 1872) Sertdartlla tmella (Alder, 1856) Sertularia distans Lamourous, 1816 Sertularia perpusilla Stechow, 1919 Sertularia marginata (Kirchenpauer, 1864) Sertularia turbinata (Lamouroux, 1816) Thyroscyfihusfruticesus (Esper, 1793) Syntheciidae 1: 1 g Splhecium tvansi {Ellis & Solander, 1786) Suborder Proboscoida Superfamily Carnpanulariidea Campanulariidae 24: 13 m, 10 g, 1 mg Campanularia hincksii Alder, 1856 Campanulaha uolubifis (L,,1758) Ctytia discoidea (Mayer, 1900) Ctytia gracilis (Sars, 1851) Clylia hemisphanica (L., 1767) Qtia linearis (Thornely, 1899) Clytia mccraityi (Brooks, 1888) (*) Chtia macrogonia Bouillon, 1984 Ctytia noliformis (McCrady, 1859) Clytiapaultasis (Vanhoffen, 1910) Ctytia pentala (Mayer, 1900) Chtia serrulata (Bale, 1888) Gonolhyraea h e n i (Allman, 1859) Hartlaubella gelatinosa (Pallas, 1776) Laomedea anguiaia Hincks, 1861 Lomedea calceoliftra (Hincks, 1871) C B TA B C CT CT IP TA C CT g K m? m? m rn? m? m? m? rn? rn ? IP m? C B TA TA B K g g MEDITERRANEAN HYDROMEDUSAE h m e d e a flexwsa Alder, I856 Laomedea neglecta Alder, 1856 Obelia bidentata Clarke, 1875 Obelia dickotomu (I.., 1 758) * Obeliajimbriata (Dalyell, 1848) Obelia geniculata (I.., 1758) Obelia longissima (Pallas, 1766) O r f h o f i s agmmetrica (Stechow, 1919) Orlkolyxis crmata (Hartlaub, 1901) O&opyxts Integra (Macgillivray, 1842) Order LAINGIOMEDUSAE 1: 1 m Laingiidae 1: 1 m Kantiella enigmatics Bouillon, 1978 Order LIMNOMEDL'SAE 9: 7 rn, 1 mg, 1 g Armohydridae 1: 1 rng Armhohydra janowzczi Swedrnark & 'I'eisrier, 1958 Microhydrulidae 1: 1 g Microh~drdapontica Valkanov, 1965 Olindiasidae 6: 6 m Gospedamia sowerbyi Lankester, 1880 Gonionemus vertens A. Agassiz, 1862 Gossea corynetes (Gosse, 1853) Matotias inexpecida Ostrournoff, 1896 Olindias phehorica (Delle Chiaje, 1841) Scolionma suvaensf (Agassiz & Mayer, 1899) C CT MA I'A TA IP m m rn m rn rn F'roboscidactylidae 1: 1 rn Proboscidactyia omata (McCrady, 1857) Order NARCOMEDUSAE 20: 20 m Aeginidae 2: 2 rn Aqinia citrea Eschscholtz, 1829 Solmundella bitentaculata (Quay & Gaimard, 1833) Cuninidae 9: 9 m Cuninafrugfera Kramp, 1948 Cunina globosa Eschscholtz, 1829 Cunina laiiventris Gegenbaur, 1856 (doubtful, probably C. globosa) Cunina octonaria McCrad y, 1857 Cunina polyfiinia (Haeckel, 1879) (doubtful) Cmina poboscidea E. & L. Metschnikoff, 1871 Cunina vitrea Gegenbaur, 1856 (doubtful, probably C. pruboscidea} Solmissus albescew (Gegenbaur, 1856) Solmissus incise (Fewkes, 1886) Solmarisidae 9: 9 m Pegantha mollicina (Forskal, 1775) (doubtful) Pegantha ntbiginosa (Koelliker, 1853) Pegantha triloba Haeckel, 1879 Pegantha mnaria (Haeckel, 1879) (doubtful) Solmaris corona (Keferstein & Ehlers, 1861) Solmaris flavescens (Koelliker, 1853) Solmaris leucostyla (Will, 1844) Solmaris sofma~is(Gegenbaur, 1856) Solmaris vanhoeffeni Neppi & Stiasny, 1911 C CT err CT E E E IP CT m rn rn rn m m m m m Order TRACHYMEDUSAE 17: 17 m Geryoniidae 2: 2 m Gerysnia poboscidatis (Forskal, 1775) Liriofu fetraphylla (Chamisso & Eysenhardt, 1821) err CT m rn F. BOER0 AND J. BOUILLON 266 Halicreatidae 1: 1 m ffaltscera tonzca Vanhoffrn, 1W2 * I Hahtrrphes m a w Bigelow, 1909 Pcta$idac 1: 1 rn 5'e~asus a t m Haeckcl, 1879 Ptychogastriidae 1: 1 m Pwhogastna asternides (Haeckcl, 1879) Rhopalonematidae 12 12 m Aglaura hern~~turna Peron & Liqueur, 1810 Amphagonu puiilla Harttaub, 1909 Ar;ta.podfina ampla (Varihof1en, 1902) Hornwowma plv&snoa Browne, 1903 Panthachqon hatckelt Maas 1893 Panthiichigan &tare (Maas, 1893) P e r a 7ncoIarata M~Crddy,185J Ransonta krumpi ,Ranson, 1932) Rhopalonema funfranurn Vanhoffcn, 1902 Rhopalonema velaturn Gegenb'~ur, 1856 Smmikea euygastra Gegenbaur, 1856 'letrorcht~eythrngaster Bigrlow , 1909 Order ACTINULIDAE 3: 3 g Halarnmohydridae 2: 2 g Halammuhydra vct@odides Remane, 1927 Halamohydra sh&ez Remane, 1927 Otohydridac 1: 1 g Otuh$ra vagans Swedmark & Teissier, 1958