Spatial Distribution of Medusa Cunina octonaria and Frequency of Parasitic Association with Liriope tetraphylla (Cnidaria: Hydrozoa: Trachylina) in Temperate Southwestern Atlantic Waters

© 2020 Academia Sinica, Taiwan Open Access Spatial Distribution of Medusa Cunina octonaria and Frequency of Parasitic Association with Liriope tetraphylla (Cnidaria: Hydrozoa: Trachylina) in Temperate Southwestern Atlantic Waters Francisco Alejandro Puente-Tapia1,*, Florencia Castiglioni2, Gabriela Failla Siquier2, and Gabriel Genzano1 1Instituto de Investigaciones Marinas y Costeras (IIMyC), Facultad de Ciencias Exactas y Naturales, Universidad Nacional de Mar del Plata (FCEyN, UNMdP), Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET), Mar del Plata, Argentina. *Correspondence: E-mail: [email protected] (Puente-Tapia). Tel +5492236938742. E-mail: [email protected] (Genzano) 2Laboratotio de Zoología de Invertebrados, Departamento de Biología Animal, Facultad de Ciencias, Universidad de la República, Montevideo, Uruguay. E-mail: [email protected] (Castiglioni); [email protected] (Siquier) Received 16 April 2020 / Accepted 27 August 2020 / Published 19 November 2020 Communicated by Ruiji Machida This study examined the spatial distribution of the medusae phase of Cunina octonaria (Narcomedusae) in temperate Southwestern Atlantic waters using a total of 3,288 zooplankton lots collected along the Uruguayan and Argentine waters (34–56°S), which were placed in the Medusae collection of the Universidad Nacional de Mar del Plata, Argentina. In addition, we reported the peculiar parasitic association between two hydrozoan species: the polypoid phase (stolon and medusoid buds) of C. octonaria (parasite) and the free-swimming medusa of Liriope tetraphylla (Limnomedusae) (host) over a one-year sampling period (February 2014 to March 2015) in the coasts of Mar del Plata, Argentina. We examined the seasonality, prevalence, and intensity of parasitic infection. Metadata associated with the medusa collection was also used to map areas of seasonality where such association was observed. Cunina octonaria was found from southern Uruguay to the coast of Mar del Plata (34.8–38.2°S, 57.2– 54.0°W), with the highest abundances and frequency of occurrence in the Río de la Plata estuary. The parasitic association was identified from the austral warm period (spring-summer season) until midautumn. Out of the 21,734 L. tetraphylla specimens that were examined, 316 were parasitized (prevalence = 1.5%) exclusively in the manubrium and gastric peduncle, with an infection intensity of 1 to 2 stolons per host. Furthermore, the medusoid buds per stolon ranged from 11 and 29 at different stages of development. No significant differences were observed between the umbrella diameter of parasitized and non-parasitized L. tetraphylla specimens, nor was any significant correlation identified between umbrella diameter and prevalence, and intensity of infection. According to the aggregation coefficient, C. octonaria had an overdispersed distribution in the host population. All parasitized hosts showed stomach vacuity due to the location of the stolon, which blocked the mouth of the host. We identified the parasitic association in the coasts of Mar del Plata, as well as in both coasts of the Río de la Plata Estuary (UruguayanArgentinean coasts). In the Southwestern Atlantic, several biological interactions between medusae and other groups have been identified; however, the specific host selectivity of C. octonaria for L. tetraphylla was not previously identified. Here we discuss the ecological importance of this association during the Citation: Puente-Tapia FA, Castiglioni F, Siquier GF, Genzano G. 2020. Spatial distribution of medusa Cunina octonaria and frequency of parasitic association with Liriope tetraphylla (Cnidaria: Hydrozoa: Trachylina) in temperate southwestern Atlantic waters. Zool Stud 59:57. doi:10.6620/ ZS.2020.59-57. Zoological Studies 59:57 (2020) doi:10.6620/ZS.2020.59-57 1 © 2020 Academia Sinica, Taiwan BACKGROUND Medusae (i.e., Medusozoa, which includes Hydro-, Scypho-, Cubo-, and Stauromedusae) (see Marques and Collins 2004; Lewis Ames 2018) are an abundant group distributed in a wide spectrum of aquatic ecosystems, including marine-coastal environments (Larson 1986; Suchman and Sullivan 2000) and some continental freshwaters systems (Arbačiauskas and Lesutienė 2005) occurring in all latitudes (Mills 1995), from polar to tropical waters, whereas in the water column they can be found from the surface to abyssal depths (Vinogradov 1968), including benthic and planktonic environments (Miranda et al. 2016). One of the ecological roles of this group includes a great variety of biological interactions with several groups such as protists, cnidarians, trematodes, mollusks, crustaceans, echinoderms, fish, among others (Ohtsuka et al. 2009 and references therein). The study of the medusae community from temperate Southwestern Atlantic waters (SWAw) (southern Brazil, Uruguay, and Argentina) has allowed us to identify the hydromedusae faunal composition, the spatial distribution (mainly of the most abundant species) (Genzano et al. 2008; Oliveira et al. 2016; Dutto et al. 2019), and some of the above mentioned interactions (e.g., Sal Moyano et al. 2012; Díaz Briz et al. 2012 2017). However, knowledge of the spatial distribution of some species, such as Cunina octonaria McCrady, 1959 (Trachylina: Narcomedusae: Cuninidae), is scarce. The present study represents the first attempt to describe C. octonaria’s spatial distribution along this vast area of the SWAw, as well as the association between this species and Liriope tetraphylla (Chamisso and Eysenhardt, 1821) (Trachylina: Limnomedusae: Geryoniidae), which has not been previously reported in SWAw. We highlight this particular parasitic relationship because the two species of hydromedusae are involved, each one in a specific stage of its life cycle: L. tetraphylla acts as host during its freeswimming medusa phase and C. octonaria acts as a parasite during its polypoid phase (stolon and medusa buds). In addition, although the polypoid phase and the free-swimming medusa of C. octonaria are the same species, each one has different reproductive strategies and energy requirements. Hydrozoans are a group of cnidarians with complex and diverse life cycles, as well as reproductive strategies that result in a wide range of diversity of polyps, colonies, and medusa morphologies, as well as complete loss or reduction of the polyp or medusa stage in some species (Cartwright and Nawrocki 2010). Morphological and molecular data have allowed us to classify to Hydrozoa into two sister groups: Hydroidolina and Trachylina (Bouillon et al. 2006). Trachylina includes the most enigmatic cnidarians, as they differ from other hydrozoans in their morphology, development, and life cycles (Collins et al. 2008). This group includes four orders: Limnomedusae, Narcomedusae, Trachymedusae, and Actinulida. Order Limnomedusae is a small group with a dimorphic benthic-pelagic cycle, characterized by a small or even nonexistent polyp in some species (Bouillon et al. 2004). When there is no polyp, its embryo directly transforms into the medusa (e.g., L. tetraphylla) (Osadchenko and Kraus 2018). For Narcomedusae species, development is indirect and does not include a real intermediate sessile polyp stage (Mayer 1910); their connection to organic, benthic substrates is lost and their entire life cycle occurs in the pelagic realm (Bouillon 1987; Bentlage et al. 2018). However, there is another way to establish an association with substrates: by using another living organism during a larval stage (Bouillon 1987), which may develop in a sessile stage that lives in parasitic association with other cnidarians (e.g., C. octonaria) (Bouillon 1987; Osborn 2000). This type of association may be necessary for some species at a certain phase of their life-cycles and may, or may not, present host specificity. The larval stages of several Narcomedusae (such as C. octonaria) parasitize the gastrovascular system and manubrium of other medusae, using their host’s food and tissues (Boero and Bouillon 2005; Raskoff and Robison 2005). Once grown, they leave their hosts and live freely (Boero and Bouillon 2005). Although medusae have a wide range of morphologies and complex life history patterns (Bouillon and Boero 2000), few published studies holoplanktonic life history of the narcomedusae. Additionally, we report the southern limit of the spatial distribution of this particular parasitic association in the Southwestern Atlantic, thus increasing the knowledge of biological associations of gelatinous zooplankton (Cnidaria and Ctenophora) on Uruguayan and Argentinean coasts. Key words: Argentine Sea, Stolon phase, Medusoid buds, Limnomedusae, Narcomedusae. page 2 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan describe the association between hydromedusae and other cnidarians. The study of biological associations has implications for marine biodiversity (Purcell and Arai 2001). Forming an association with medusae can enhance the population or diversity of the commensals, since medusae provide shelter, act as food (Masuda et al. 2008), and protect the larval stages from predators (Ohtsuka et al. 2009). Analyzing this intragroup association (i.e., intrahydrozoan) revealed aspects of its population dynamic, life history, life cycle variation (Benovic et al. 1987; Giagrande et al. 1994; Lucas and Reed 2009), and ecological role (e.g., trophic and ecological niches) (Marcogliese 2005). These aspects are essential to understanding how populations develop and persist through the year (Graham et al. 2001; Lucas and Reed 2009). Parasites may be useful indicators of food-web structure and function because they depend on the presence of other hosts in their life cycles for transmission (Marcogliese 2005). There are numerous advantages of using parasites as indicators of trophic processes and food structure; for example, identifying direct trophic links between the host and other organisms, host trophic level, and predators (Marcogliese 2005). MATERIALS AND METHODS Spatial distribution of Cunina octonaria To analyze the historical distribution of C. octonaria in temperate SWAw, we examined 3,288 zooplankton samples of the Medusae collection (MedCol) of the J.J. Nágera Coastal Station, Universidad Nacional de Mar del Plata (UNMdP), Argentina were revised. The collection includes zooplankton samples taken over 36 years (from 1980 to 2016) along the Uruguayan and Argentine continental shelves (34–56°S, 52–70°W), comprising an area ⁓7 million km2 (Fig. 1). The sampling was done using fishery research vessels operated by the Instituto Nacional de Investigaciones y Desarrollo Pesquero (INIDEP, Argentina), as well as local research programs designed by the Instituto de Investigaciones Marinas y Costeras (IIMyC-CONICET, Mar del Plata, Argentina) and the Instituto Argentino de Oceanografía (IADOCONICET/UNS, Bahía Blanca, Argentina). A variety of plankton nets (Bongo, Nackthai, Motoda, Pairovet, Calvet, and Multinet) were used in oblique trawls from the proximity of the bottom to the surface. The medusae were sorted, counted, and identified, and the abundances were calculated and expressed as ind.m3 of filtered waters. Cunina octonaria occurrence was plotted based on its presence/absence in the set of samples. The spatial distribution of L. tetraphylla has been previously described in Dutto et al. (2019). To analyze and compare the possible differences between austral warm (spring-summer) and cold periods (autumn-winter) regarding the frequency of occurrence and abundances of C. octonaria along with the geographical distribution, the dataset was divided into two periods: October to March (spring-summer) and April to September (autumn-winter). Then, we plotted the presence/absence data and abundances of both periods; we performed Pearson’s chi-square test to test the possible differences between both periods and Student t-test for the abundances differences (Sokal and Rohl 1999). Parasitic association While analyzing the temporal variation in the gelatinous zooplankton species (GZ) species in the Mar del Plata Harbor (Argentina) (38°08'17"S–57°31'18"W) (Fig. 1), we identified the parasitic association between the polypoid phase of C. octonaria and the freeswimming medusa of L. tetraphylla; therefore, we used these samplings to described the seasonality of this association. The zooplankton sampling was performed over a one-year sampling period (February 2014 to March 2015). The samples were collected with weekly or biweekly frequency during the warm period and monthly frequency during the cold period with oblique tows using a standard zooplankton net (mouth diameter: 75 cm; mesh size: 500-μm) and a flowmeter attached to the net mouth, which allowed us to calculate the volume of filtered water to estimate the numerical abundance (ind.m3) of the GZ. We first analyzed this parasitic association by describing the temporal variation in the abundances of the involved species. For L. tetraphylla, we reviewed the previous reports of Puente-Tapia and Genzano (2019), and for C. octonaria we carried out several analyses: a one-way ANOVA to test if abundances of the medusa phase were significantly different between seasons. If significant differences were present, then a post hoc Tukey’s HDS test was performed to examine which seasons showed differences (Zar 1999). Data were previously normalized by logarithmic transformation to fulfill the assumption of homogeneity of variances (Levene test). The analyzed specimens were stored at the MedCol with the tag Cunina-Liriope-001). The total number of parasitized organisms per sample was recorded, as were the number of stolons on each host specimens. With these data, the A) parasitic prevalence and B) intensity of infection were described. We determined these infection levels first in terms of the total values (i.e., considering the period of occurrence page 3 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan of the association) and later for each month. According to Bush et al. (1997): A) prevalence is the number of hosts infected with one or more C. octonaria polypoid structures per sample divided by the number of hosts and expressed as a percentage; B) intensity of infection is the number of stolons of C. octonaria in a single infected host. In addition, we recorded the number of medusoid buds of C. octonaria in each stolon. The total umbrellar diameter (UD, in mm) of the hosts was measured under a stereomicroscope to determine the relationship between host size and parasite load (considering only the number of stolons). Thus, the specimens were immersed in water to adopt a similar position that is observed in the marine Fig. 1. Map of the study area: temperate Southwestern Atlantic waters. Yellow circle indicates the area studied for seasonality variation (Mar del Plata Harbor, Argentina, 38°08'17"S-57°31'18"W); red crosses represent the zooplankton samples of the dataset from the Medusae collection. RPE (Rio de la Plata estuary); BsAs (Buenos Aires); SB (Samborombón Bay); SCT (San Clemente del Tuyú); MdP (Mar del Plata); BB (Bahía Blanca); SMG (San Matías Gulf); VP (Valdés Peninsula); SJG (San Jorge Gulf); TdF (Tierra del Fuego); M/F (Malvinas/Falkland Islands). The fine lines represent the 50, 100, 200, and 1000 m isobaths in the region. Scale bar = 250 km. N page 4 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan environment. A general size-frequency histogram per host was constructed to analyze the occurrence of the different host sizes with the presence/absence of polypoid phase of C. octonaria. The length classes were arbitrarily defined by applying the Sturges’ rule (Zar 1999). Possible differences between the sizes of parasitized and non-parasitized hosts were analyzed by a Student’s t-test (Sokal and Rohlf 1999). After checking for the normality of the data distribution and homogeneity of variance, we used Pearson’s correlation coefficient (r) (Zar 1999) to analyze the relationship between the host size and prevalence and intensity of infection for each size group. After calculating the coefficient (r) and evaluating its significance, a Student’s t-distribution test was applied. Considering the number of medusoid buds per host and using the r-coefficient, we examined the correlation of each size group. The level of statistical significance was P ≤ 0.05 for all analyses. The assumption of normality of data (Kolmogorov-Smirnov test) and homogeneity of variance (Levene test) were verified using the previous normalization with logarithmic transformation of the data (Zar 1999). To estimate the distribution of C. octonaria in the host population, we calculated the aggregation coefficient (k) of the negative binomial distribution. This analysis determined whether the host species had a normal, random or overdispersed distribution (Morales and Pino 1987). The concept of host range (Rohde 2005) was used to classify parasitic species as specialists or generalists. According to this concept, a specialist parasite species has a marked affinity for a specific host family, genus or species, while a generalist parasitizes hosts from a number of different taxa. For this analysis, we reviewed the available literature on the associations between genus Cunina and other hosts species. We examined the gut content of the all L. tetraphylla organisms (including parasitized and non-parasitized) in terms of vacuity (i.e., presence or absence of stomach content), which could be an indicator of the negative effects C. octonaria caused L. tetraphylla. Subsequently, the metadata associated with the MedCol served to determine A) the areas of spatial co-occurrence of L. tetraphylla and C. octonaria and B) geographical areas and the seasonality where this parasitic association was observed. RESULTS Spatial distribution of Cunina octonaria in temperate Southwestern Atlantic waters According to the data from the MedCol, this narcomedusa was found in 28 of 3,288 zooplankton samples (0.85%). It had a spatial distribution from 34.8–38.2°S and 57.5–54.0°W (i.e., from southern Uruguayan coasts to waters of central-northern Buenos Aires Province, Argentina) (Fig. 2A). The main occurrence was observed in the estuarine area of the Río de la Plata. All samples showing C. octonaria specimens were collected at a depth lower than 50 m. Although, 92.85% of these samples (n = 26) were collected during the warm period and only 7.14% (n = 2) during the cold period, no significant differences were observed (x2 = 2.0, d.f. = 1, P = 0.16) regarding the frequency of occurrence between both periods. During the cold period, the specimens were collected only in San Clemente del Tuyú (⁓36°S–56.8°W) and Mar del Plata coasts (⁓38°S–57.5°W), with abundances of 0.04 to 1.08 ind.m3 (Fig. 3A). During the warm period, the abundances ranged between 0.003 and 28.05 ind. m3, showing the highest abundances ⁓130 km south of the Samborombón Bay, followed by the Río de la Plata estuarine area (close to Uruguay) (Fig. 3B). No significant differences were found in abundances between the warm and cold periods (t = 0.89, d.f. = 24, P = 0.38). Liriope tetraphylla showed a southern limit up to ⁓38°50'–40°S, ⁓60–61°W (⁓Bahía Blanca estuary, the adjacent shelf of El Rincón and Monte Hermoso, Buenos Aires Province, Argentina) (Fig. 2B) (see Dutto et al. 2019). Analysis of parasitic association In samples collected over the year (February 2014 to March 2015) in Mar del Plata Harbor, the free-swimming medusa of C. octonaria was only observed during the spring and summer seasons, with abundances ranging from 0.003 (October) to 3.82 ind. m3 (February) (Table 1). In terms of seasonality, no significant differences were observed in the abundance values (ANOVA: F = 1.62; P = 0.34). Cunina octonaria was found in salinities ranging from 33.7 to 36, while the water temperature oscillated between 14 and 22.3°C. Liriope tetraphylla occurred year-round, with the highest abundances occurring during the warm period and the lowest in the cold seasons. Their monthly abundances ranged from 0.01 ind.m3 (May) to 80.98 ind.m3 (March) (Table 1). Significant differences were observed between climate seasons (ANOVA: F = 14.30, P = 0.01), particularly between autumn-spring (Tukey’s HSD test = -1.70, P = 0.01), autumn-summer (Tukey = -2.28, P < 0.01), and winter-summer (Tukey’s HSD test = -1.50, P = 0.01), i.e., between the cold and warm periods. A total of 18 gelatinous species were identified (14 hydromedusae, 1 scyphomedusae, and 3 ctenophores), of which the polypoid phase of C. octonaria was page 5 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan only observed in association with L. tetraphylla. The parasites were observed during the spring and summer seasons, with water temperatures ranging from 14.0 to 22.0°C, and salinity from 34.0 to 36.0. Of the 21,734 L. tetraphylla individuals analyzed during the period of co-occurrence, 316 were infected (total prevalence = 1.45%). All hosts had a single stolon structure (intensity of infection) with medusoid buds, except for one that had two stolons. The number of medusoid buds per host ranged from 11 (December) to 29 (January). These values varied monthly (Table 2). However, the umbrellar diameter (UD) of L. tetraphylla ranged from 0.5 to 14.3 mm, parasitized specimens oscillated from 1.7 to 7.0 mm. Medusae between 5 and 6 mm of the size-frequency histogram showed the highest number of parasitized specimens, with prevalence values of 2.5 and 3.5%, respectively (Fig. 4). No significant differences were observed between the UD of parasitized and non-parasitized specimens (t = 1.0, d.f. = 281, P = 0.30). No significant correlations between the UD and prevalence were observed (r = 0.02, n = 244, P = 0.72), nor between UD Fig. 2. Spatial distribution in temperate Southwestern Atlantic of the free-swimming medusa phase of Cunina octonaria (A) and Liriope tetraphylla (B) (including parasitized and non-parasitized specimens). Orange cycles represent the samples of zooplankton with species of medusa; black crosses represent the lack of individuals. (C) Areas of the association between the polypoid phase of C. octonaria and L. tetraphylla (red cycles). Acronyms in figure 1. Scale bars = 250 km. (A) (B) (C) page 6 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan and intensity of infection (number of stolons). However, a significantly positive correlation between the number of medusoid buds per host and the UD was observed (r = 0.61, n = 73, P < 0.01, t = 0.16, d.f. = 71, P < 0.05): i.e., the number of medusoid buds increased as the UD of the L. tetraphylla increased. The host species showed a parasite overdispersed distribution according to the aggregation coefficient (k = 0.32). That is, most of the examined organisms harbored few or no parasites. In contrast, several host individuals contained a large number of this narcomedusa. Observations in vivo showed that the stolon of C. octonaria had several medusoid buds in different growth stages; they protruded from the characteristic manubrium and gastric peduncle of L. tetraphylla. A section of the stolon was inside the manubrium and Fig. 3. Comparison of the spatial distribution of the abundances (ind.m3) of the free-swimming medusa phase of Cunina octonaria between the (A) austral cold (autumn-winter) and (B) warm periods (spring-summer) in temperate Southwestern Atlantic waters. Red arrows (2A) indicate the regions where in which abundances were recorded during the cold period. Acronyms in figure 1. Scale bars = 250 km. (A) (B) Table 1. Abundance values (ind.m3) of the free-swimming medusa phase of Cunina octonaria and Liriope tetraphylla during the study year (February 2014 to March 2015) in Mar del Plata Harbor, Argentina Cunina octonaria Liriope tetraphylla Water temperature (°C) Salinity 2014 February 0.29 10.90 22.0 35.0 March 0.08 80.98 19.1 35.0 April 0.04 0.27 15.3 35.0 May 0 0.01 14.2 35.0 June 0 0.08 11.7 34.0 July 0 0.41 10.4 35.0 August 0 1.45 9.8 35.0 September 0 0.26 11.8 34.0 October 0.003 4.22 14.0 33.7 November 0 0.73 16.3 36.0 December 1.16 19.53 18.8 34.0 2015 January 0.73 20.99 19.3 35.3 February 3.82 5.22 21.2 36.0 March 0 19.95 22.3 36.0 page 7 of 15Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan another one on the outside (Fig. 5A–D). All stolon structures had medusoid buds. The stolon had an elongated and cylindrical shape widening in the basal region to the interior of the manubrium, while the external side had flat or oval shapes (Fig. 5C–D). Newly liberated medusae and free-swimming medusa of C. octonaria are shown in figure 5E–F. In addition to the Mar del Plata, we identified the association in San Clemente del Tuyú (Argentina) (36°21'S–56°43'W), but Castiglioni and Failla Siquier (personal observations) also identified it in Maldonado, Uruguay (34°38'S–54°5'W). These findings indicate that the association is present off both coasts of the Río de la Plata estuarine area (Uruguay-Argentina) (Fig. 2C). On Mar del Plata coasts, the association was observed in the austral spring (October and December) and summer (from January to March), while in San Clemente del Tuyú, it was observed during the mid-autumn (mid-April and mid-May). Finally, on the Uruguayan coasts, it was identified from January to March (summer to early autumn). A general analysis of the gut content of all individuals of L. tetraphylla when associations occurred showed that the stomachs of the parasitized individuals were completely empty, whereas some of the nonparasitized organisms had chaetognaths, crustaceans, and detritus. This vacuity might be due to the location of the polypoid phase, which prevented food intake. DISCUSSION Spatial distribution of Cunina octonaria and Liriope tetraphylla in temperate Southwestern Atlantic waters Cunina octonaria is a Narcomedusae widely distributed in the Pacific, Atlantic, and Indian Oceans (Burke 1975; Segura-Puertas 1984; Galea 2007; Segura-Puertas et al. 2009; Andrade Ruíz 2010; Oliveira et al. 2016; Schuchert 2020). In the SWAw it has been identified from Colombia to Argentina (Oliveira et al. 2016). In the present study, we described the distribution in temperate waters, going from the southern Uruguayan to Mar del Plata coasts (⁓38°S) as the southern limit, with the higher abundances and frequency of occurrence in the Río de la Plata estuary and adjacent waters. However, according to Zamponi (1983), isolated records have shown its presence near the Malvinas/Falkland Islands (⁓51°46'S–59°31'W). Although we reviewed more than 3,200 zooplankton samples, ⁓20% (n = 647) of which were distributed under this latitude, C. octonaria was not found in the samples from the region indicated by Zamponi (1983); therefore, his finding could not be confirmed. During the study year in Mar del Plata, C. octonaria was found in temperatures between 14 and 22.3°C. According to Soares Moreira (1978), in laboratory conditions, C. octonaria shows little Table 2. Period of co-occurrence and parasitological parameters (prevalence and intensity of infection) of the association between Liriope tetraphylla (Lt) and Cunina octonaria (Cu) at Mar del Plata Harbor, including measurements of physicochemical and biological parameters Period of cooccurrence Total number of Lt analyzed Total number of Lt parasitized Prevalence (%) Intensity of infection Number of medusoid buds of Cu Min-Max Surface water temperature (°C) Salinity February 2014 336 7 2.1 1 11-14 22.0 35.0 March 2014 14,757 44 0.3 1 12-21 19.1 35.0 October 2014 1,925 2 0.1 1 11-13 14.0 33.7 December 2014 549 1 0.2 1 12-18 18.8 34.0 January 2015 4,167 262 6.3 1-2 13-39 21.2 36.0 Total 21,734 316 1.5 2 Fig. 4. Percentage of parasitized and non-parasitized Liriope tetraphylla for the different size classes. Grey bars: pooled data of non-parasitized hosts; black bars: specimens parasitized with polypoid phase of Cunina octonaria. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 1-3 4-6 7-9 10-12 13-15 Frequency (%) Host umbrellar diameter (mm) Non-parasitizedParasitized page 8 of 15 Zoological Studies 59:57 (2020) © 2020 Academia Sinica, Taiwan tolerance to low temperatures, and grows ideally at 15°C. At 10°C the specimens crumped, while at 5°C they became completely quiet. Therefore, it is possible that the Patagonian and Sub-Antarctic regions are unfavorable zone for C. octonaria due to their low temperatures, mainly during the austral cold period. During the cold period, the shelf circulation of the temperate SWA consists of the northward flow cold waters of the Malvinas/Falkland Current, which is characterized by low-salinity and nutrientrich waters. In contrast, during the warm period, the influence is from the southward tropical-subtropical flow warm waters of the Brazil Current, characterized by oligotrophic and salty waters (Piola and Matano 2001; Piola et al. 2010). According to Vannucci (1957) and Navas-Pereira (1973), C. octonaria is a typical species in the Brazil Current. Earlier records confirm this to be a widely distributed frequent species in warm waters (Kramp 1965); consequently, the fact that this narcomedusa was only found during the warm period and sporadically at the early autumn could be due to passive transportation through the warm current. In this context, the contribution of the Río de la Plata waters are a natural barrier to the distribution of different marine organisms (Boltovskoy et al. 1999). However, this barrier is intermittent because, in periods of low flow and with the effects of northeastern winds, it is possible to find species that usually live in northern areas of Argentina (Uruguay and Brazil), in southern regions and localities such as Mar del Plata (Mianzan et al. 2001; Mianzan and Acha 2008). On the other hand, several studies have focused on L. tetraphylla because it is the most frequent taxon in temperate SWAw (Dutto et al. 2019). This species was mainly found in the Río de la Plata estuary and surrounding areas (Fig. 4B). During the study year in Mar del Plata, L. tetraphylla was observed in temperatures between 9.8 and 22.3°C and salinity ranging from 33.7 to 36 (Puente-Tapia and Genzano 2019). According to Vannucci (1957), this species showed optimal temperatures from 20 to 25°C and 35–36 of salinity. However, it was observed in strongly Fig. 5. Parasitic association between the polypoid phase of Cunina octonaria (1) (parasite) and Liriope tetraphylla (host) from Mar del Plata, Argentina. (A-D) Location of the stolon in the manubrium and gastric peduncle of L. tetraphylla; (D) detail of the stolon and the medusoid buds; (E) newly liberated medusae from the stolon; (F) comparison between newly liberated medusa (2) and adult individual (3) of C. octonaria. Lowercase letters indicate the structure of the host species: m (manubrium); mo (mouth); su (subumbrella); t (tentacles); eu (exumbrella). Scale bars: A–C and F = 1 mm; D–E = 500 μm. page 9 of 15 Zoological Studies 59:57 (2020)