MARINE ECOLOGY PROGRESS SERIES Vol. 168: 147-162,1998 Published July 9 Mar Ecol Prog Ser

Production, respiration, and photophysiology of the Cassiopeaxamachana symbiotic with : effect of jellyfish size and season

E. Alan verde1r', L. R. McCloskey2

'Oregon State University. Department of Zoology. 3029 Cordley Hall, Corvallis, Oregon 97331. USA *~aSierra University, Department of Biology, 4700 Pierce Street, Riverside, California 92515, USA

ABSTRACT: The association between the symbiont microadriaticum (zooxanthellae) and its jellyfish, xamachana, was investigated as a function of jellyfish size and season. Symbiont cell diameter and volume were higher dunng January than September. Although zooxan- thella-specific chlorophyll was independent of jellyfish size, both chlorophyll a and c were higher dur- ing January. Regardless of season, algal density and jellyfish size were inversely related. The die1 mitotic index (MI) of zooxanthellae was phased, with a peak of 0.25 y: occurring between 09:OO and 12:OO h. September photosynthetic rates were always higher than January rates and reflected the sea- sonal light and temperature regimes at the latitude of the Florida Keys (USA). , when normalized to either zooxanthella density or protein, displayed an inverse relationship with jellyfish slie ?v?~ii~l~dliIB~~II~LIUII 1die3 dlbu biluwed iuve~be~eidii~llsilip w~ii~ jeiiyiisil size, w~iilSepielll'uer metabolism being higher than that of January. The carbon budgets calculated for these rnedusae indi- cate that the carbon photosynthetically fixed by the zooxanthellae, and subsequently translocated to the host, is capable of satisfying about 169% of the host's metabolic demand (CZAR) and is indepen- dent of both jellyfish size and season. These seasonally influenced physiological effects underscore the necessity for seasonal examinations of algal-cnidarian syrnbioses in order to understand the photo- physiology of the association on an annual basis

KEY WORDS: - Zooxanthellae . Jellyfish . Photosynthesis . Respiration . Photo- physiology. Carbon budgets. CZAR

INTRODUCTION of zooxanthellae. Differences in the excretion rates of ammonia between aposymbiotic and symbiotic Cassio- The relationship between marine unialgal symbionts pea sp. suggest that the zooxanthellae remove host- (zooxanthellae) and cnidarian hosts has been more produced ammonia and aid in the recycling of nitrogen widely studied for benthic cnidarians (reviews by Mus- (Cates & McLaughlin 1976). Hofmann & Kremer (1981) catine 1990, Battey 1992, Davles 1992, Trench 1993) and Hofmann et al. (1996) investigated the importance than for motile ones. Early studies on the scyphozoan of zooxanthellae to strobilation in Cassiopea androm- Cassiopea sp. showed that endosymbiotic zooxanthel- eda and reported that the presence of algae was indis- lae (Symbiodinium) are capable of CO2 uptake and fix- pensable although algal photosynthesis or photosyn- ation (Balderston & Claus 1969, Drew 1972) and subse- thate release was not essential for this process. quent carbon translocation to host tissues (Balderston Apparently, the symbiotic zooxanthellae produce and & Claus 1969). Cates (1975) suggested that these pho- supply a 'factor' which promotes the elevated metabo- tosynthesis-respiration interactions enhance the eco- lism required for proceeding through developmental logical efficiency of Cassiopea sp. due to the presence stages (Rahat & Hofmann 1987). The photosynthetic carbon budget of the epipelagic coronate medusa Linuche unguiculata has been inves-

0 Inter-Research 1998 Resale of full article not permitted 148 Mar Ecol Prog Ser 168: 147-162. 1998

tigated by Kremer et al. (1990); zooxanthellae from Algal parameters. Chlorophylls: After homogeniza- these medusae are capable of contributing carbon tion of each jellyfish, the algae were obtained by 3 cen- products towards host respiratory requirements trifugation~,each at 3 to 4 min at top speed in a clinical (CZAR) of approximately 160%. A similar study on the centrifuge, with the algal pellet resuspended each time swimming rhizostomid medusa Mastigias sp. was con- in filtered seawater. Three replicate aliquots (3.0 m1 ducted by McCloskey et al. (1994); CZAR estimates each) of resuspended algae were processed as averaged 97 and 176% for lake and lagoon ecotypes, described by Verde & McCloskey (1996). Between 15 respectively. In contrast to both L. unguiculata and and 20 hemacytometer grids were counted to obtain Mastigias sp., Cassiopea xamachana is more seden- algal density (cells ml-l) of the suspension which was tary, spending most of its time with the bell's exum- used for chlorophyll assays. The chlorophyll content brella in direct contact with the substrate and its tenta- was calculated using the equations of Jeffrey & cles elevated towards the surface (Blanquet & Phelan Humphrey (1975) as moMied by Parsons et al. (1984). 1987).Therefore, we wanted to investigate whether or Density, size, and biomass: Algal numbers from not these semi-sedentary medusae rely as much on 20 replicate samples from each homogenate were symbiont-derived organic products for metabolism as counted with a hemacytometer and converted to total do L. unguiculata and Mastigias sp. standing stock of algae per jellyfish. Concurrently, the To our knowledge, no study estimating the photo- diameters of 20 randomly chosen zooxanthellae were synthetic capabhties and resultant carbon budgets of measured from each jellyfish homogenate. For biomass scyphozoans has investigated these parameters as a studies, freshly collected, jellyfish (n = 4, January and function of seasonal environmental changes. Most September) were individually homogenized in filtered photosynthesis-respiration estimates are based on a seawater. Aliquots of the algal suspension were cen- single-case time period, as is the case for Linuche trifuged for 3 to 4 min in a clinical centrifuge and the jel- unguiculata (Kremer et al. 1990), Mastigias sp. lyfish tissue layer above the algal pellet was discarded. (McCloskey et al. 1994), and Cassiopea xamachana Jellyfish contam~nati.on,of the algal suspension was pe- (Vodenichar 1995). The few seasonal-based studies on riodically checked by hernacytometer counts; after the scyphozoans have only dealt with jellyfish population alga1:nematocyst ratio exceeded 95 %, 20 hemacytome- dynamics or distributions (Brewer & Feingold 1991, ter grids were counted to obtain algal density (cetlsml-l). Olesen et al. 1994) or growth (Olesen et al. 1994). With Samples for C:N analysis were prepared for isolated respect to medusan size, 2 studies on Mastigias sp. are algae as follows. Either 2.0 m1 or 3.0 m1 of the jellyfish- the only investigations that have documented the free algal cell suspension was filtered, under gentle influence of jellyfish size on photosynthesis, respira- vacuum, through a precombusted (450°C for 12 h) tion, and carbon budgets (McCloskey et al. 1994) and GF/C filter (25 mm Whatman). Each filter was subse- algal symbiont population density and regulation quently rinsed with 0.5 m1 of hstdled water, to remove (Muscatine et al. 1986). Consequently, we also wanted salts, and frozen. Four seawater-filter blanks were to determine if medusan size or seasonal environmen- included in all sample sets as controls. Prior to C:N tal change affects the carbon budgets of C. xamachana analysis, the filters were dried (45°C for 12 h) and the symbiotic with Sym biodinium microadriaticum (Freud- carbon and nitrogen masses quantified using a Control enthal 1962, Trench & Blank 1987). Equipment Corporation Elemental Analyzer (Model 240X). After accounting for seawater blanks, the algal carbon and nitrogen masses (pg cell-') were calcu- MATERIALS AND METHODS lated. The nitrogen mass was multiplied by 6.25 to obtain protein per algal cell (Muscatine et al. 1986). Collection and processing of jellyfish. Different-sized Mitotic index and growth: Tentacle snips chosen specimens of Cassiopea xamachana were collected ddy at random from freshly collected and individually from a shallow, mangrove-lined lagoon at Little Conch marked medusae (January and September) were col- Key (mile marker 62.3) in the Florida Keys (USA). All lected every hour for 24 h and frozen. Each sample was production and respiration experiments were conducted homogenized and the dividing algal cells counted in a during September 1992 and January 1993. After each hemacytometer. The mitotic index (MI) was calculated experiment, bell diameter, wet weight (ww), and dis- as described by Wilkerson et al. (1983). The phased- placement volume of each jellyfish were measured. The division formulae of Vaulot (1992) was used to calcu- jellyfish were individually homogenized in filtered sea- late the algal-specific growth (p) water (Whatman no. 5 filter), the homogenate volume determined, and sub-samples taken for various assays. Each jellyfish was assigned to 1 of 3 size classes based on where f,,, and E,,,, are the maximum and minimum bell diameters: small, medium, or large. daily fraction of dividing cells, respectively. Verde & McCloskey: Photobiology of the mangrove jellyfish 149

Jellyfish and algal biomass. Frozen aliquots of the RESULTS jellyfish homogenate were thawed and the protein content of the medusan fraction estimated by the BCA Environmental parameters Protein Assay (Pierce Chemical) using bovine serum albumen (fraction V) as the standard. Total jellyfish Average (k SD) daylength during January (11.2 * biomass was determined as medusan and algal protein 0 1 h, n = 14) was significantly shorter (t-test, p < 0.001) fractions combined. Individual medusan and algal bio- than during September (12.8 * 0.1 h, n = 10). Average masses, expressed as fractions of total protein, are daily integrated irradiance during January (31.0 * referred to as Pand l-P, respectively (Muscatine et al. 9.6 E m-2d-') was significantly lower (t-test, p i 0.001) 1981). than during September (46.0 % 9.2 E m? d-l). Likewise, Photosynthesis and respiration parameters. The diel water temperature during January was significantly oxygen fluxes of intact jellyfish were measured with a lower than during September. Average (+ SD) January self-contained underwater respirometer (McCloskey water temperature (24.2 2 1.4"C, n = 14) during the et al. 1985) at a depth of 2.0 2 0.5 m. Simultaneously, night (when respiration was measured) was signifi- ambient light intensity was measured with a Li-Cor cantly lower (t-test, p 0.001) than for September quantum sensor (Model LI-192s) From data obtained (29.9 * 0.5"C, n = 10). froin each 24 h record of oxygen flux, the integrated total daily jellyfish respiration and gross photosyn- thetic rate were determined as outlined by McCloskey Algal parameters et al. (1994) and Verde & McCloskey (1996). Morning photosynthesis and irradiance (P-I) response curves Cell size. Since no significant differences (ANOVA, were generated from diel oxygen flux measurements. p > 0.05) in algal diameters as a function of jellyfish For P-I analysis, hourly net photosynthesis was normal- size during either January or September occurred, the ized to algal density and the curves were iteratively fit cell diameters from the 3 jellyfish sizes were pooled via the hyperbolic tangent function (Jassby & Platt together. The average (* SD) algal cell diameter for 1976). January was 9.08 pm (* 0.87, n = 540) and was signifi- Carbon budgets. The carbon budget parameters of cantly higher (t-test, p < 0.001) than the average of

&i55i xam;c,5;n; ..v--- ..-'-m Q1 Gpliea ---'-".-,.-l Q CA >>m,A..' Ir fl n Q7nl fnr ;.lrr;rl cnllc dup:,:nn Coptern- c IIIUULIIC.U UJIllY the V."' ,i V."'., " - V'", -v- -AY-' C-..-. 3 -- nomenclature and equations of McCloskey et al. ber. It follows that the average cell volume for algae in (1994).The contribution of carbon from zooxanthellae January of 400.6 pm3 (& 106.8, n = 540) was signifi- to respiration (CZAR) was calculated as cantly higher (t-test, p < 0.001) than the average vol- described by McCloskey et al. (1994) and Verde & ume of 347.7 pm3 (* 103.8, n = 820) for algae in Sep- McCloskey (1996). tember. Statistical analysis. Prior to analysis, all data sets Chlorophyll. Zooxanthella chlorophylls a and c (chl a were tested for compliance with the assun~ptionsof and chl c) were significantly higher during January each statistical test. If assumptions were violated, the than September (Table 1, Fig. lA, B).However, there data were transformed and retested, or non-parametric was no significant difference in symbiont chlorophyll tests were utilized. All percentage data (i.e. the M1 per cell as a function of jellyfish size. The chl a:chl c data) were arcsine transformed before analysis (Sokal ratio was not significantly different regardless of sea- & Rohlf 1995, Zar 1996). The software packages Statis- son or jellyfish size (Table l, Fig. 1C). tica\\Vm (Statsoft, Inc.) and Slidewrite Plus (Advanced Density and biomass. Since no significant differ- Graphics Software) were used to conduct statistical ences between algal densities occurred during Janu- and graphics analysis, respectively. ary and September, the data were pooled together

Table 1.Symbiodinium microadriat~curn. Mean (* SD) chlorophyll content (chl; pg cell-') and chl a:chl c ratio of zooxanthellae isolated from the jellyfish Cassiopea xarnachana during September 1992 (n = 27) and January 1993 (n = 40). The Student's t-test was used to determine significance between January and September values. ns: not significant (p > 0.05), "'p < 0.001

Chlorophyll January September Significance Jan:Sep ratio

Chl a 2 21 k 0.35 Chl c 0 66 * 0.10 Chl a + c 2.87i 0.43 Chl a chl c ratio 3.37 i 0.29 150 Mar Ecol Prog Ser 168. 147-162, 1998

D JANUARY SEPTEMBER Mitotic index and growth. An average (kSD) of 3.01 - I ?4.soh (k8.3, n = 20) of the algal symbionts resided solely within the tentacles, with th.e remaining 25.5% (k8.3, n = 20) inhabiting the bell. The average sym- biont M1 from the tentacles of Cassiopea xamachana was 0.047% (k0.068, n = 20) and from the bell was 0.084% (k0.095, n = 20), values which were not sig- nificantly different (paired t-test, p > 0.05). The M1 values were significantly different (ANOVA, p c 0.05) throughout a die1 period regardless of season and jel- lyfish size (Fig. 3). The near midday M1 peak was sig- nificantly higher than the rest of the day (Tukey, p < 0.05) and we interpret this as phased mitosis. The max- imum peaks of the MI, regardless of season or jellyfish size, were not significantly different (ANOVA, p > 0.05) and the grand mean (+SD)was 0.25% (k0.20, n = 72). Likewise, the M1 minimums were not significantly different (ANOVA, p > 0.05) and thc grand mean was 0.02% (*0.03, n = 72). Consequently, the algal-specific

U 2510 300' 600' 900 12'00 l~bd

JELLYFISHBIOMASS (rng protein) JELLYFISH BIOMASS (rng protein) Fig. 1. Symbiodinium microadriaticum and Cassiopea xam- achana. Chlorophyll content of zooxanthellae from C. xam- achana as a function of jellyfish size and season. Best-fit curve 2.68 for the relationship between chlorophyll and host biomass is linear regression. (A)Chl a, January r2 = 0 08, September r2 = 0.07; (B) chl c, January r2 = 0.03, September r' = 0.01, (C) chl a:chl cratio, combined rZ = 0.01. The slopes from each figure are not significantly different (p > 0.05) from zero and only the r2 value is listed. Sample sizes for January and September were 41 and 27, respectively

SMALL MEDIUM LARGE JE1.LYFISH DIAMETER

(Fig. 2A). Jellyfish bell diameter and algal density Fig. 2. Symbiodiniurn microadriaticum and Cassiopea xam- were inversely related, with smaller jellyfish contain- achana. Density of zooxanthellae within C. xamachana as a ing significantly higher algal cell densities than larger function of jellyfishI size and season. Best-fit curve for the rela- jellyfish (Fig. 2B).With regard to algal biomass, both tionship between algal density and host biomass (A) is a the carbon content and C:N ratio of zooxanthellae power function: Density = 4.82 (~iomassf-nl', r' = 0.64. For the relationship between algal density and host bell hameter (B) were significantly higher during January than Septem- horizontal lines connect algal densities that are not signifl- ber (Table 2), but the nitrogen and resultant protein cantly different (Tukey, p > 0.05); lines at different levels sig- masses were not significantly different regardless of nify densities that are significantly different (Tukey, p < 0.05). season. The pooled averages (kSD) for January and Values above error bars are the mean [density) for each bell size class and values within bars represent sample size. Error September nitrogen and protein values were 15.23 pg bars show i SD. Bell diameter (cm,E * SDI of each size class, cell-' (i 1.63, n = 8) and 95.16 pg cell-' (k 10.20, n = 8), regardless of season, was: Small, 4.0 i 1.1; Medium, 8.0 * 0.8; respectively. Large, 13.0 * 2.2 Verde & hdccloskey: Photobiology of the mangrove jellyfish 151

Table 2. Symbiodinium microadriaticum. Mean (+SDI carbon, nitrogen, and protein contents (pg cell-') and C:N ratio of zooxanthellae isolated from the jellyfish Cassiopca xamachana during September 1992 (n = 4) and January 1993 (n = 4).The Mann-Whitney U-test was used to determine significance between January and September values. ns: not significant (p > 0.05), 'p < 0.05

Algal parameter January September Significance Jan:Sep ratio

Carbon 114.88 + 9.40 85.29 + 9.27 1.34 N~trogen 15.89 + 1 83 14.56 + 1.30 ns 1.09 Prote~n 99.34 + 11.43 90.98 + 8.11 ns 1.09 C.N ratio 7.30 * 1.05 5.87 + 0.49 1.24

JANUARY P SEPTEMBER l7 JANUARY SEPTEMBER II $4,

,1 111, 1 -1- -1 - 300 1

L -.~. , . , , , ,A < , , <.l-.;.,.,.L..,. : : 8, U 111 12 ld It, !S-- 28s :! 24 2 4 v fi Ill l: JJ :,, IN28, L: 2, TIME OF DAY (h) Fig. 3. Symbiodinium microadriaticum and Cassiopea xam- achana. Die1 mitotic index of zooxanthellae within C. xam- achana as a function of jellyfish size (small: SML, medium: MED, and large: LRG) and season. Sample sizes for each hour, regard.less of season, for SI\4L, MED, and LRG were 10. 16, and 10, respectively. Error bars show & SD. Bell diameter (cm, X i SD) of each slze class, regardless of season, was: Small, 4.0 i 1.1; Medium, 8 0 + 0 8; Large, 13.0 + 2.2. Hori- ZOII~CI~lines denote occurrence of night

growth rate (p), regardless of season and jellyfish size, JELLYFISH BIOMASS (rng protein) was 0.0023 d-l. Photosynthesis. Daily gross photosynthesis, normal- Fig. 4. Symbiodinium microadriaticum and Cassiopea xam- ized to cell density, protein, and chl a, is shown in achana. Gross photosynthesis (P,) of zooxanthellae within C. xamachana as a function of jellyfish size and season. Best-fit Fig. 4. Cell- and protein-specific photosynthesis dur- curve for the relationship between gross photosynthesis and ing September were significantly higher than during host biomass is linear regression. (A)Density; for January r2 = January (Fig. 4A, B) and showed a significant host- 0.07, for September P, = -0.10 (Biomass) + 228.41, r' -0.27. size-specific relationship, with the photosynthesis of (B) Protein; for January r' = 0.07, for September P, = smaller jellyfish being significantly higher than that -0.99 (Biomass) + 2400.26, r2 = 0.27. (C) Chl a; for January r2 = 0.01, for September r2 = 0.07. Regression profiles where of larger individuals. The lower January rates, in con- regression equations are given display slopes significantly trast, exhibited no significant host size relationship different from zero. Sample sizes for January and September (regression, p > 0.05) and averaged (*SD, n = 27) were 41 and 27, respectively 152 Mar Ecol Prog Ser 168: 147-162.1998

+W - 600 .; .; 500 JANUARY @ SEPTEMBER E JANUARY SEPTEMBER -: 400 A g P300 A < & 200 a 2 2 2 I00 lx 300 600 900 l200 1500 ' 0 300 600 900 1200 1500 JELLYFISH BIOMASS (mg protcin) JELLYFISH BIOMASS (ing protein)

U JANUARY SEPTEMBER U JANUARY SEPTEMBER -----

z -" 300 0 M t* F: E -: 700. Mu-.< E 0" 2 F LOO- =- t

SMALL MEDIUM LARGE SMALL MEDIUM LARGE JELLYFISH DIAMETER JELLYFISH DIAMETER

Fig. 5. Symbiodinium rnicroadnahcum and Cassiopea xam- Fig. 6. Cassiopea xamachana. Respiration rate of C. xam- achana. Photosynthetic efficiencies (a) of zooxanthellae achana as a function of jellyfish size and season. Best-fit curve within C. xamachana as a function of jellyfish size and season. for the relationship between jellyfish respiration and biomass Best-fit curve for the relationship between a and host biomass (A) is a power function: January resp. = 607.91 (Bi~mass)"'.~~, (A) is a power function: January a = 0.08(Bioma~s)-~.~~,r2 = 9 = 0.74; September resp. = 409.75 (Bioma~s)-O.'~,r2 = 0.38. 0.32, September a = O.lO(Bi~massf-~-'~,9 = 0.44. For the rela- For the relationship between respiration and host bell diame- tionship between cr.and host bell diameter (B]horizontal lines ter (B) honzontal lines connect respiratlon rates that are not connect a values that are not significantly different (Tukey, significantly different (Tukey, p > 0.05), and lines at different p > 0.05), and lines at different levels signify values that are levels signify rates that are significantly different (Tukey, p < significantly different (Tukey, p < 0.05). Student's t-test: ns, 0.05). Student's t-test: ns, not significant; "p < 0.01; "'p < not significant; 'p < 0.05. Error bars show * SD Values within 0.001. Error bars show + SD. Values within bars represent bars represent sample size. Bell diameter (cm, 2 i SD) of sample slze. Bell diameter (cm, F * SD) of each size class, each size class, regardless of season, was: Small, 4.0 * 1.1; regardless of season, was: Small, 4.0 * 1.1; Medium, 8.0 +- 0.8; Medium, 8.0 i 0.8; Large, 13.0 i 2.2 Large, 13.0 * 2.2

152.64 + 36.57 pg O2 d-l X 10-6 cells and 1604.01 * cases, average January a were higher than September 384.26 yg O2 d-l mg-l protein, respectively. Chloro- a although they were only significantly higher for phyll-specific photosynthesis during September was medium and large jellyfish. Regardless of season, also significantly higher than photosynthesis during small jellyfish showed significantly higher a. than large January (Fig. 4C). However, neither January nor jellyfish. September photosynthetic rates related to jellyfish size. The average (+SD) January and September photosynthetic rates were 70.69 (k21.93, n = 27) and Jellyfish respiration 142.35 (k34.43, n = 40) pg 0, d-' ~ig-'chl a, respec- tively. Jellyfish respiration rates durlng September were January-acclimated jellyfish displayed significantly significantly higher (ANCOVA, p < 0.01) than during higher photosynthetic efficiencies (a) (ANCOVA, p < January (Fig. 6A), particularly for medium and large 0.01) than their counterparts during September jellyfish (Fig. 6B). Regardless of season, smaller jelly- (Fig. 5A). Jellyflsh size-specific a displayed both sea- fish displayed significantly higher respiration rates sonal and size dependent differences (Fig. 5B). In all than larger individuals. Verde & VcCluskey: Photobiology of the mangrove jellyfish 153

U JAVtlARY m SEPTEMBER / nl

SO[ r2 = 0.96 o,,;,' l 7a /

JELLYFISH BIOMASS (~ngprote~n) JELLYFISH BIOMASS (~ngprotein)

Flg. 7 Cassiopea xarnacliana. Daily carbon budget parameters of C xamachana as a function of jellyfish size and season. Best- fit curves for the relationship between carbon budget parameter and jellyfish biomass are linear regression. (A) Net photosyn- thesis (P,): January P,, = 0.054 (Biomass) + 1.64, r' = 0.92; September P, = 0.074 (Biomass) + 2.49, r2 = 0.96. (B)Algal carbon-spe- cific growth rate (C,): January C,, = 0.00032 (Biomass) + 0.011, r2 = 0.97; September C, = 0.00027 (Biomass) + 0.0047, r2 = 0.98. (C)Potentially translocated carbon (C,): January C, = 0.054 (Biomass) + 1.63, r' = 0.92; September C, = 0.073 (Biomass) + 2.49, r2 = 0.96. (D)Animal respiratiun (R,): January R, = 0.032 (Biomass) + 1.38, rL = 0 97; September R, = 0.043 (Biomass) + 1.94, r' = 0.90. (E)Available translocated carbon in excess of that utilized for animal respiration (C,,,.,,,): C.,,.,,,,= 0.022 (Biomass) + 1.09, r' = 0.63. (F) Potential carbon contribution of zooxanthellae towards the animal's respiratory requirements (CZAR): CZAR = 0.0050 (Biomass) + 171.19, r'= 0.001. Sample sizes for January and September were 41 and 27, respect~vely

Carbon budgets (ANCOVA, p < 0.001) with jellyfish size (Fig. ?C) and, as for P,, September C, rates were significantly higher Daily net photosynthesis (P,,), algal carbon-speciflc (ANCOVA, p < 0.05) than Jan-uary C, rates. growth rate (C,), translocated carbon (C,),animal res- Host R, was directly related (ANCOVA, p .c 0.05) to piration (R,), carbon available for host growth (Cava,,), biomass (Fig. ?D)with R, dunng September being sig- and CZAR are presented in Fig. 7.Algal P, per jellyfish nificantly higher (ANCOVA, p < 0.01) than in January. showed a direct relationship (ANCOVA, p < 0.001) Like the other parameters, Cc,radwas also drectly related with jellyfish biomass (Fig ?A), with September rates (ANCOVA, p < 0.05) to jellyfish biomass (Fig. ?E);how- significantly higher (ANCOVA, p < 0.001) than in ever, there was no significant difference (ANCOVA, p > January (Fig. ?A). C, also displayed a direct relation- 0.05) between January and September values. In con- ship (ANCOVA, p < 0.001) with jellyfish biomass (Fig. trast to the other carbon budget parameters, CZAR did 7B). However, January C,, was significantly higher not show a significant trend (ANCOVA, p > 0.05) with (ANCOVA, p < 0.05) than September C,. The calcu- either season or jellyfish size (Fig. ?F). The average lated C, (i.e. P,, - C,,) also showed a direct relationship (+ SD) CZAR for all jellyfish was 169.2% (* 65.2, n = 68). 154 Mar Ecol Prog Ser 168: 147-162, 1998

The distribution of carbon for an average Cassiopea meters on both the photobiology and basic physiologi- medusa (5 800 mg protein) acclimated to January or cal aspects of algal-cnidarian symbioses. Both the pho- September conditions is shown in Fig. 8A, B, respec- tosynthetic (and resultant carbon budgets) and meta- tlvely. With the exception of C,, carbon budget para- bolic rates of Cassiopea xamachana are seasonally meters (either estimated or directly measured) were dependent, with higher rates occurring during Sep- between 39 and 48% higher for September- than Jan- tember (summer) than January (winter), and are con- uary-acclimated jellyfish. In contrast, January C, esti- tingent on both light and thermal regimes. Winter- mates were 20 % higher than those in September. acclimated symbionts photoadapt to reduced light conditions by increasing their internal pools of chloro- phyll and becoming photosynthetically more efficient. DISCUSSION However, these compensatory mechanisms are not enough to enhance photosynthetic rates to the levels The principal conclusion of this study concerns the observed during the summer. Such seasonally depen- definitive influence of seasonal environmental para- dent physiological effects emphasize the requirement for seasonal investigations of algal-cnidarian symbioses in order to comprehend the ecological physiology and photobiology of the symbiotic _---_ asscciation on an annual cycle. / / / / Algal parameters / / I Cell size. A shift in algal cell size occu.rred as a l New Animal Biomas3 function of season for Cassiopea xamachana, I with medusae acclimated to 'winter' environmen- I

I tal conditions in the subtropical Florida Keys har- l boring larger-sized zooxanthellae (9.08 pm diam- \ L eter) than jellyfish acclimated to summer \ conditions (8.64 pm diameter). Whether this phe- \ nomenon is an adaptation to environmental con- \ ditions or is directly regulated by the jellyfish , . host is not known, but it seems reasonable to posit that larger algal size may increase light- p, = ? gathering efficiency. A. January Cussiopeuxurncccl~anu Muscatine et al. (1.986) reported an average cell diameter for zooxanthellae from small and large Mastigias sp. of 9.85 and 8.25 pm, respectively, while Kremer et al. (1990) reported a diameter of ----_ 8.9 pm from Linuche unguiculata These values , are similar to those we measured for zooxanthel- / / / lae from Cassiopea xamachana. / / I Scsc Anim:~lBiom;~?;~ I Fig. 8. Symbiodinium microadriaticum and Cassiopea I I xamachana. Average carbon fluxes for zooxanthellae from C. xamachana during (A)January and (B)Septem- ber. Values are given as mg C d-l for an average slzed jellyfish which is I800mg protein. P, and P, are gross and net photosynthesls, respectively; R, and R, are algal and animal respiration, respectively; C, is the carbon- specific growth rate; C, is the carbon translocated to the host; p, is expelled algae; C,,,, is the carbon available to the animal for mucus production, storage products, growth, and reproduction; CBAG is carbon back-trans- located from the animal to the algae. Sample sizes for B. September C'a~siopeuxalnuc~buna January and September were 31 and 27, respectively Verde & McCloskey: Photobiolc )gy of the mangrove jellyfish 155

Chlorophyll. Vodenichar (1995) reported an average value of 2.8 X 106 cells mg-' protein. This is almost symbiont chl a content of approximately 1.0 pg cell-' identical to the algal density of small-sized C. xam- during summer conditions, which is lower than our val- achana but is 1.7 times higher than that for medium- ues of 1.45 pg cell-' for September zooxanthellae. The sized medusae. Clearly there are significant differ- average algal chl a of zooxanthellae (during January) ences as to how different of symbiotic jellyfish from Cassiopea xamachana (2.21 pg cell-') was similar maintain their complen~ent of zooxanthella symbionts to the 2.1 and 2.0 pg cell-' reported by Kremer et al. with respect to body size, but there are no obvious pat- (1990) and Wilkerson & Kremer (1992) for Linuche terns or trends yet evident among the 3 species for unguiculata, and the 2.0 pg cell-' for Mastigias sp. which we have comparative data. reported by McCloskey et al. (1994). Likewise, the Regulation of algal symbionts by Cassiopea xam- mean chl a:chl c rat10 for algae in C. xamachana (3.37) achana seems to occur primarily by host digestion of was similar to the average ratios for algae from L. algae and by expulsion into the environment. Fitt & unguiculata (3.23). Trench (1983) showed that heat-killed algae which In Cassiopea xarnachana, the zooxanthella chloro- underwent phagocytosis were susceptible to lysosomal phyll content during January was 1.5 times higher than attack by acid phosphatase. This suggests that endo- for algae during September. Since both the daylength symbiotic algae that become senescent, for whatever and light intensities were significantly lower during reason, may be consumed by the host through phago- January than in September, these differences in irradi- cytocis. Individual jellyfish, maintained in buckets with ance levels may be the reason winter-acclimated zoo- seawater, secreted mucus bands containing many xanthellae photoactively increase their internal pools zooxanthellae (authors' pers. obs.), although the of chlorophyll, as described for both symbiotic and expelled zooxanthella numbers were not quantified. free-living (Richardson et al. 1983, Whether C. xamachana expels zooxanthellae continu- Prezelin 1987). Alternatively, there may just be an ously, at times when internal pools of zooxanthellae increase in syrnbiont chlorophyll content concomitant become too high or when the animal host is stressed, is with larger cell volumes. not currently known. Like C. xarnachana, large-sized Density. Vodenichar (1995) showed an average algal Mastigias sp, seems to regulate algal populations density of approximately 4.5 X 106 cells g-' ww for Cas- (Muscatine et al. 1986) by expulsion or digestion. How- sicpea x;m;c,5';;na ranging ir, size from ? te !C! cm he!! ever, in sma!!er-sized Mastig~ar;

numbers reported by Muscatine et al. (1986) and Kre- in other soft-bodied invertebrates (Schlichter 1984, mer et al. (1990). DeFreese & Clark t991, Preston 1993) and are subse- Kremer et al. (1990) and Wilkerson & Kremer (1992) quently util.ized by the algae to support phased divi- reported that algae from Linuche unguiculata have a sion. Such uptake of carbon from decomposing man- nitrogen content of 13.3 to 15.7 pg N cell-', which is grove litter is also expected to be higher during the very similar to values reported in this study. The pro- summer than winter due to higher decomposition rates tein content of zooxanthellae from Cassiopea xarn- that occur during the summer season (Mackey & Smail achana is 2.5 times higher than that for algae from 1996). Alternatively, inorganic nitrogen in the form of Mastigias sp., which have an average of 36.3 pg pro- ammonium may also be directly absorbed transepider- tein cell-' (Muscatine et al. 1986). Muscatine et al. mally from the water, much as for Mastigias sp. (Mus- (1986) and Kremer et al. (1990) reported C:N values of catine & Marian 1982) or L~nucheu.nguiculata (Wilker- 9.7 and 6.2, respectively, which are also similar to the son & Kremer 1992), and from the sediment, since the C:N ratio of zooxanthellae from C. xamachana. In sum- concentrations of these nutrients are higher inshore mary, the biomass parameters of Symbiodinium micro- than offshore (Szmant & Forrester 1996). Similarly, dis- adriaticum within C. xamachana are nearly identical to solved organic matter may also be absorbed and used those of zooxanthellae from other jellyfish. as a food source, as is the case for Heteroxenia Mitotic index and algal growth. Wilkerson et al. fuscescens (Schlichter 1982a, b, Schlichter et al. 1983, (1983) reported that zooxanthellae from Mastigias sp. 1984). display a phased, diel cell division profile (sensu Alternatively, high levels of by the host on McDuff & Chisholm 1982). Wllkerson et al. (1983) zooplankton could support phased algal div~sion.Cas- attributed such a phased M1 to exposure of the algal siopea xamachana is a rapacious predator of Artemia symbionts to a 2 h pulse of ammonium at night. Other nauplii during feeding studies in aauaria [E.A.V per<. cnidarians exhibiting phased symbiont division in- obs.). Support for prey-enhanced algal symbiont clude the Seriatopora hystrix (Hoegh-Guldberg growth is provided by McAuley & Cook (1994), who & Smith 1989), Stylophora pistillata, Fungia repanda, reported that zooxanthellae in starved colonies of the and Pocillopora damicornis (Smith & Hoegh-Guld- hydroid Myrionema amboinense show an M1 of 3.9%, berg 1987). Clifford & Blanquet (1991) also reported whereas in colonies fed Artemia nauplii, the M1 that zooxanthellae extracted from Cassiopea xam- increased by a factor of 2.6 to 10.08%. McAuley & achana and cultured with a 14 h 1ight:lO h dark cycle Cook (1994) also provide evidence that external n~tro- display a diel pattern of cell motility, with phased gen sources directly influence zooxanthella division division occurring during the latter part of the dark rates. Other studies corre1atin.g increased zooxanthel- period. lae cell division with host feeding include Fitt & Cook As we have shown, zooxanthellae in Cassiopea xam- (1990) on M. amboinense and Cook et al. (1988) on the achana also appear to display a phased algal MI. In anemone Aiptasia pallida. contrast to the findings of Clifford & Blanquet (1991), From the algal-specific in hospite growth rate value (p) the zooxanthellae wlthln C. xamachana demonstrated of 0.0023 d-' for zooxanthellae in Cassiopea xamachana, a daytime phased d~visionpattern between 09:OO and one can compute a doubling time for the symbiont 12:OO h. Why in vitro phased division occurs in the dark popul.ation, of 301 d using the equation, of Wilkerson et al. and in hospite phased d~visionoccurs during daytime [1983; D, = (In 2)p-']. W~thsuch, low growth rates and is not known, but one possibility may be that some long doubling times, it is unlikely overgrowth by the host-derived fraction regulates algal cell division. symbionts poses a threat to the host. The average p of Precedence for such regulation is provided by Carroll zooxanthellae from Mastigiassp. and Linuche unguicu- & Blanquet (1984a, b) and Blanquet et al. (1988) for lata is 0.10 d-' and 0.02 to 0.04 d-', respectively (Wilker- alanine uptake and by McDermott & Blanquet (1991) son et al. 1983, Kremer et al. 1990, McCloskey et al. for glucose uptake by zooxanthellae via inhibitory 1994). However, these relatively high growth rates are fractions derived from host homogenates. based on a duration of cytokinesis of 11.0 h (Wllkerson et Unlike Mastigias sp., Cassiopea xamachana does not al. 1983) whereas the growth rate estimate of zooxan- exhibit diel migration but remains stationary on the thellae from C. xamachana does not utilize this number sediment-water interface. It may be that carbon and since the diel division rates are phased. The p of nitrogen from bacterial metabolism at the sediment 0.0023 d-' for symbionts from C. xamachana is closer to surface or tentacles (Schiller & Herndl 1989) or from those measured for zooxanthellae from the corals Pocil- mangrove litter decomposition (Lugo & Snedaker lopora eydouxi (0.0017 d-', Davies 1984) and Pontes 1974, Fell et al. 1975, Newel1 et al. 1984, Robertson pontes (0.0023 to 0.0028 d-', Edmunds & Davies 1986, 1988, Mackey & Smail 1996) are transepidermally 1989) and the temperate anemone Anthopleura elegan- absorbed by C, xarnachana as are various amino acids tissima (0.0029 d-', Verde & McCloskey 1996). Verde & McCloskey: Photobiology of the mangrove jellyfish 157

Table 3. Mean photosynthesis versus irradiance (P-I) parameters of 3 jellyfish. Cassiopea xamachana,Mastigias sp. and Linuche unguiculata.The P-I data of Mastigias sp. and L. unguiculata were obtained from McCloskey et al. (1994) and Kremer et al. (1990), respectively. The C. xamachana data were obtained from small and medium sized jellyfish from both January and Sep- tember months which corresponded to the sizes of Mastigjas sp and L, ungujculata.C:h4 = C. xanlachana Mastigjas sp ratlo. C:L= C. xarnachana:L. unguiculata ratio. na: not available

P-Id Cassiopea Nfastigias Linuche C:M C:L

a 1, lop,

1P,,' rnax

dP-I parameters: a is a measure of photosynthetic eff~ciency[(pg O2h-' X 10.' cells) (pE m-' S-')-'] Ikis the intersection of a and P," max tangents (PE m-' S-') I,,,, is the optimum irradiance during P," max (pE m-%-') P," max is the maximum net photosynthesis (pg 0' h-' X 10-= cells)

I b~herea = 0.09 mg 0, h-' mg-' chl a and chl a cell' = 2.1 pg (I 0 mg chl a = 476.2 X 10' cells) Therefore, (0.09mg O2 h-' mg' chl a) (476.2 X 10%ells)-' = 0.189 [(pgO2 h-' X 10-6 cells) (PE m-2 S-')-']

'Where P," max = 4.5 mg O2 h'' mg' chl a and chl a cell.' = 2.1 pg (1.0 mg chl a = 476.2 X 10"ells). Therefore, (4.5 mg O2 h-' mg-' chl a) (476.2 X 10' cells)-' = 9.5 (pg O2 h-' X 10-" cells)

The determination of the M1 only from symbionts synthetic rates (and resultant carbon flux values) specifically located within the tentacles may be justi- would be expected to be even higher than reported fiably questioned, since the algal M1 from the tenta- here for September. On the other hand, January-accli- cles of some cnidarians is elevated compared to the mated symbionts have significantly higher cr and rest of the organism. Muller-Parker (1987) reported chlorophyll content, suggesting that these algae adapt an ;lg;! hlI fro% !he tentacles cf .4!ptzsia p~!rhe!!at~ to these lower light and temperature regimes by be 1.5 times higher than the M1 from the anemone being more efficient at capturing incident light. Simi- column. In contrast, a tentacle-based M1 for either lar seasonal photosynthetic rates were reported by zooxanthellae or zoochlorellae from Anthopleura ele- Barnes (1988) on the (Davis Reef, gantissima is similar to the M1 of symbionts from the Australia), where gross photosynthesis was more than whole anemone (E.A.V. unpubl. data). For Cassiopea 30 % higher during August compared to December. xamachana, we deemed it appropriate to determine When a 10% reduction in respiration during Decem- algal growth rates from tentacle-derived symbionts ber was considered, the res'ultant net photosynthesis because the majority of algae (75%) were located was higher by 45% (Barnes 1988).The conclusions of within the tentacles and this value was similar to the Barnes (1988) and those reported in this study under- ?O'",b reported for zooxanthellae in Mastigias sp. score the necessity for seasonal studies on algal- (Muscatine & Marian 1982). These tentacle snips also cnidarian symbioses. permitted time-course evaluations of M1 changes in When photosynthesis is examined as a function of single jellyfish. For these reasons, we have used the jellyfish size, the highest rates occurred in smaller- more conveniently-obtained M1 values from tentacle sized individuals. This scaling of photosynthetic rates snips. is due to the significantly higher algal densities and Photosynthesis. Although their photosynthesis-irra- metabolic rates of smaller jellyfish, both of which pro- diance parameters differ (Table 3) the photosynthetic mote elevated photosynthesis. Similar results were rates of Casslopea xamachana and other symbiotic obtained by Fisher et al. (1985) from the Tri- jellyfish, such as Mastigias sp., Linuche unguiculata, dacna gigas and Vodenichar (1995) in Cassiopea xam- and , are very similar (Table 4). achana in which both photosynthesis and respiration Not surprisingly, a strong seasonal influence on pho- displayed an inverse relationship with animal size. tosynthetic rates takes place. Since light intensity, The reduction of photosynthetic rates with increas- light duration, and water temperature are signifi- ing jellyfish size during September may be due to algal cantly greater during September than during January, shading, either from increased algal self-shading or higher photosynthetic rates would be expected during from medusan pigmentation. The idea that algal self- September. If summer measurements were conducted shading plays a greater role in reducing photosynthe- during June (i.e. summer solstice), the summer photo- sis than host pigmentation in Cassiopea xamachana is 158 Mar Ecol Prog Ser 168: 147-162, 1998

Table 4. Comparison of the average maximum gross photosynthetic rates of 4 tropical symbiotic jellyfish: Mastigias sp., Linuche unguiculata, Cassiopeaandromeda, and Cassiopea xarnachana.na: not available

Jellyfish PgOmax P," max Source (pgO2 h-' X 10-' cells) (pg O2 h-' pg-' chl a)

Mastjgiassp. Lake 17.6 8.8 McCloskey et al. (1994)d Lagoon 29.6 14.8 Lin uche unguiculata 12.6 6.0 Kremer et al. (1990)" Cassiopea andronleda na 3.7-4.2 Hofmann & Kremer (1981)" Cassiopea xamachana January 13.6 6.3 Present study September 15.8' 11.1 Present study

dFrom Table 7 of McCloskey et al. (1994) "From Table 4 of Kremer et al. (1990) =Based on a grand mean regardless of jellyfish size -

supported by the work of Blanquet & Phelan (1987). Jellyfish respiration They found that the unique mesogleal polymeric gly- coprotein pigment 'Cassio Blue' in C, xamachana It is difficult to directly compare the respiration rates attenuated harmful solar radiation while permitting of Cassiopea xamachana with other jellyfish metabolic the penetration of photosynthetically active radiation. rates (Kremer et al. 1990, McCloskey et al. 1994) due to Consequently, in this instance, this particular host the problem of normalizing units and size variation. pigment does not significantly reduce algal photosyn- Small-sized jellyfish display higher mass-specific thesis. metabolic rates than larger individuals and this inverse Since Cassiopea sp. are generally found in shallow scaling between animal size and metabolism is a well- depths (Drew 1972) ranging from 0.1 m (seagrass and known biological phenomenon. Similar size-specific mangrove beds) to 10 m ( borrow pits, see Clark & metabolic rates were reported by Vodenichar (1995) for DeFreese 1987) in Florida, they are exposed to high C. xamachana. September metabolism is much higher light intensities and ultraviolet (UV) radiation. The than that of January, and this is probably a simple con- symbiont photosynthetic rates are concomitantly sequence of elevated temperature effects since the av- higher without showing any photoinhibition effects erage water temperature during September (29.g°C) (authors' unpubl. data) and these jellyfish may protect was significantly higher than for January (24.2"C). themselves from O2 toxicity by producing both super- oxide dismutase (SOD) and catalase (Dykens 1984, Lesser 1989, Lesser & Shick 1989a, b). Similarly, they Carbon budgets would be expected to maintain elevated quantities of UV-absorbing compounds similar to those found The individual carbon budget components usually within the mucus of Fungia fungites (Drollet et al. displayed a direct relationship with jellyfish biomass, 1993) or rnycosporine-like amino acids (MAAs) (Shick and the 3 components P,, C,, and R, are significantly et al. 1992, 1995, Stochaj et al. 1994). Banaszak & higher during September due to higher water temper- Trench (1995a, b) described the effects of UV on C. atures, Light intensities, longer daylength, or a combi- xamachana and its symbionts. They showed that Sym- nation of all three. In contrast, January C, is signifi- biodinium microadriaticum from C. xamachana syn- cantly higher than the C,, during September, which we thesizes 3 MAAs (mycosporine-glycine, shinorine, and believe is a direct consequence of the higher carbon porphyra-334). These MAAs are subsequently translo- per algal cell during January. The C,,,d is the quantity cated to the host and are used by the host to minimize of carbon available to the animal host for reproduction, deleterious UV effects. It is these environmentally storage, growth, mucus production, or the synthesis of driven physiological adaptations which have permitted a suite of important metabolites (such as SOD, cata- this symbiotic association to thrive at high densities in lase, MAA's, and peroxidase). shallow areas of the tropical and subtropical subtidal The potential contribution of algal carbon products zone. to animal respiration (CZAR) showed no relationship Verde & McCloskey: Photobiology of the mangrove jellyfish 159

with either season or jellyfish biomass. This is counter- Four parameters were not estimated or measured in intuitive because P,,, C,, and C, showed such a strong this study of Cassiopea xamachana carbon budgets: relationship with biomass. The explanation is that, CBAG, new animal growth, px,and heterotrophy. The since R, is also directly related to biomass, CZAR val- potential for carbon to be back-translocated from the ues, regardless of season, are not correlated with jelly- animal to the algae (CBAG, see Verde & McCloskey fish size. CZAR is an estimate of the integrated PR 1996) 1s a remote possibility. Jellyfish growth rates are ratio; if net photosynthesis (minus algal carbon-specific also an unknown variable and remain to be estimated. growth) is greater than host respiration, then CZAR is The carbon expelled from the association in the form of greater than 1 (i.e. 100%). intact algal cells, p, (Hoegh-Guldberg et al. 1987, An average CZAR of 169% suggests that the zoo- McCloskey et al. 1996), is also an unknown for C. xam- xanthellae are potentially capable of providing all of the acliana. Heterotrophy represents the amount of carbon carbon necessary to support the respiration of the obtained from captured prey or dissolved organic car- medusa, and Cassiopea xamachana may not require any bon (DOC) directly absorbed by both the ectoderm and external carbon sources. This value is similar to that re- endoderm as previously discussed. ported for Linuche unguiculata (160%, Kremer et al. Another source of carbon loss from this association is 1990) and for lagoon Mastjgiassp. (143 %, McCloskey et the quantity of mucus that can be released by Cassio- al. 1994). A comparatively lower CZAR of 97 % for lake pea xamachana. Between 40 and 60 % of the daily car- A4astigias sp. was attributed to elevated host metabolism bon fixed by photosynthesis in the coral resulting from this morph's die1 migration patterns (Mc- acuminata and Acropora formosa was released as Closkey et al. 1994). A CZAR of 169% for C. xamachana mucus or DOC (Crossland 1980, Crossland et al. 1980a, may actually be a conservative estimate due to likely b). Recently, Hansson & Norrman (1995) estimated that presence of a resident microbial fauna associated with an average of 1.2 mg C d-' is released as DOC in the the tentacular region of the medusa (sensu Schiller & temperate jellyfish Aurelia aurita. Consequently, Herndl1989, Hansson & Norrman 1995). The combined mucus production and loss from C. xamachana could bacterial respiration was not factored out of the calcula- be a significant carbon sink for this association. This tion of the jellyfish respiration, which would overesti- excreted mucus is likely used to foster bacterial growth mate host metabolism, and consequently, CZAR values on the medusan tentacles (Hansson & Norrman 1995), wuuid be i-ebiicei: iii magniiiide coi.;lmensurcite vv7ithttc ar,d this micrehi~!pepu!;rti~n cni~!d he used dir~rtlyor extent of microbial metabolism. indirectly as a carbon resource as described by Schiller Vodenichar (1995) reported CZAR values of approx- & Herndl (1989) for several coral species. imately 106% in medusae with a bell diameter greater than 6 cm, whereas smaller-sized jellyfish showed CZAR values of approximately 73%. Vodenichar Acknowledgements. This research was supported by a Walt (1995) suggested that these smaller medusae as well as Disney World Company grant and Houston Underwater Club SeaSpace Scholarship (to E.A.V.) and a National Science symbiotic larvae and scyphistomae may require exter- Foundation grant OCE-8303513 (to L.R.M.). We thank the nal carbon sources to meet their respiratory carbon staff of both The Living Seas, EPCOT, Walt Disney World requirements. In contrast to the CZAR estimates of Company and Conch Key Cottages for the use of boats and Vodenichar (1995), we calculate an average CZAR facilities and the Florida Department of Natural Resources for (+SD) value of 171 % (+88, n = 22) for medusae with a collecting permits. We express our gratefulness to K. Clark for collecting jellyfish, W. Fitt for his pre-submission review of bell dlameter less than 6 cm. Because these 2 studies thls lnanuscrlpt and other remarks I-egardingCass~opea, and utilized decidedly different methods and equipment V. Weis for an additional evaluation. We thank 3 anonymous for measuring photosynthesis (with differing algal den- reviewers for their critical comments; their suggestions sities) and respiration, these differences undoubtedly greatly enhanced this manuscript. The help of A. Cleveland. K. McCloskey, and R. Verde during this project was much contributed to the respective dissimilarities in CZAR appreciated. estimates. Vodenichar (1995) measured both photo- synthesis and respiration in a laboratory environment by taking O2 readings every 5 min for a total of 30 min LITERATURE CITED at ambient light intensities above 800 pE m-' S-' and Balderston WL, Claus G (1969) A study of the symbiotic calculated CZAR assuming 8 h of saturated photosyn- relationship between Symbiodinium microadriaticum thesis. We believe that our CZAR estimates of 171% Freudenthal, a zooxanthellae and the upside down jelly- are more accurate and realistic for medusae of this size fish, Cassiopea sp. Nova Hedwigia 1?:3?3-382 since our measurements of both photosynthesis and Banaszak AT, Trench RK (1.995a) Effects of ultraviolet (UV) radiation on marine rnicroalgal-invertebrate symbioses. I respiration took place in situ over the course of 24 h at Response of algal symbionts in culture and in hospite. the site where abundant densities of Cassiopea xam- J Exp Mar Biol Ecol 194:213-232 achana were located. Banaszak AT, Trench RK (1995b) Effects of ultraviolet (UV) 160 Mar Ecol Prog Ser 168: 147-162, 1998

radiation on marine microalgal-invertebrate symbioses. 11. Aldabra Atoll. J Exp Mar Biol Ecol9:65-69 The synthesis of mycosporine-like amino acids in response Drollet JH. Glaziou P, Martin PM (1993) A study of mucus to exposure to UV in Anthopleura elegantlssima and Cas- from the solitary coral Fungia fungites [: siopea xamachana. J Exp Mar Biol Ecol 194:233-250 Fungiidae) in relation to photobiological UV adaptation. Barnes DJ (1988) Seasonality in community productivity and marBiol 115:263-266 calcification at Davies Reef, Central Great Barrier Reef. Dykens JA (1984) Enzymic defense against oxygen toxicity in Proc 6th Int Symp 2:521-527 marine cnidarians containing endosymbiotic algae. mar Battey JF (1992) Carbon metabolism In zooxanthellae-coelen- Biol Lett 5:291-301 terate symbioses. In: Reisser W (ed)Algae and symbioses. Edmunds PJ, Davics PS (1986) An energy budget for Pontes Biopress Ltd, Bristol, p 153-187 porites (Scleractinia). Mar Biol 92:339-347 Blanquet RS, Ernanuel D, Murphy TA (1988) Suppression of Edmunds PJ, Davies PS (1989) An energy budget for Porites exogenous alanine uptake in ~solatedzooxanthellae by porites (Scleractinia) growing in a stressed environment. cnidarian host homogenate fractions: species and symbio- Coral Reefs 8:37-43 sis specificity. J Exp Mar Biol Ecol 117: 1-8 Fell JW, Cefalu RC, Masters IM, Tallman AS (1975) Microbial Blanquet RS, Phelan MA (1987) An unusual blue mesogleal activity in the mangrove () Leaf detritus protein from the mangrove jellyfish Cassiopea xam- system. In: Walsh GE, Snedakar SC, Teas HJ (eds) Proc Int achana. Mar Biol94:423-430 Syrnp Biology and Management of (Honolulu, Brewer RH, Feingold JS (1991) The effect of temperature on HI). Univ of Florida. Gainesville, p 661-679 the benthic stages of Cyanea (: ), and Fisher CR, Fitt WK, Trench RK (1985) Photosynthesis and res- their seasonal distribution in the Niantic River estuary, piration in Tridacna gigas as a function of irradiance and Connecticut. J Exp Mar Biol Ecol 152:49-60 size. Biol Bull 169:230-245 Carroll S, Blanquet RS (198la) Aianine uptake by isolated Fill WK, Cook CB (1990) Some effects of host feedlng on zooxanthellae of the mangrove jellyfish, Cassiopea xam- growth of zooxanthell.ae in the marine hydroid Wlyrionema achana. I.Transport mechan~smsand utilization. Biol Bull amboinensein the laboratory and in nature. In: Nardon P, 166:409-418 Glaninazzi-Pearson V, Greni.er AM, Margulis L,Smith DC Carroll S, Blanquet RS (1984b) Alanine uptake by Isolated (eds) Endocytobiology IV. Inst~tut National de la zccx?n!he!!ae sf thc =-cgrov= jc!.!jrfish, Gissi~pe;xam- Keci-~erii~eAyiu~lullliqua, Fd~is, p 26i-264 achana. 11. Inhibition by host homogenate fraction. Biol Fitt WK, Trench RK (1983) Endocytosis of the symbiotic Bull 166:419-426 Symbiodinium mirrnadriaticum Freuden- Cates N (1975) Productivity and organic consumption in Cas- thal by endodermal cell of the scyphistomae of Cassiopeia siopea and Condylactis. J Exp Mar Biol Ecol 18:55-59 xamachana and resistance of the algae to host digestion. Cates N. McLaughlin JJA (1976) Differences of ammonia J Cell Sci 64:195-212 metabolism in symbiotic and aposymbiotic Condylactus Freudenthal HD (1962) Symbiodinium gen. nov. and Symbio- and Cassiopea spp. J Exp Mar Biol Ecol21:l-5 dinium microadriatjcum sp. nov., a zooxanthellae: taxon- Clark KB, DeFreese DE (1987) Population ecology of omy, life cycle and morphology. J Protozool9:45-52 caribbean ascoglossa (Mollusca: Opisthobranchia): a Hansson U, Norrrnan B (1995) Release of dissolved organic study of specialized algal herbivores. Am Malacol Bull 5: carbon (DOC) by the scyphozoan jellyfish Aurelia aurita 259-280 and its potential influence on the production of planktic Clifford JM, Blanquet RS (1991) Effects of host homogenate bacteria. Mar Biol 121:527-532 on the motility of endozoic algae from Cassiopea xam- Hoegh-Guldberg 0, McCloskey LR, Muscatine L (1987) achana. Am Zoo1 31 1-154 Expulsion of zooxanthellae by symbiotic cnidanans from Cook CB, D'Elia CF. Muller-Parker G (1988) Host feeding and the Red Sea. Coral Reefs 5:201-204 nutrient sufficiency for zooxanthellae in the Hoegh-Guldberg 0, Smith JS (1989) Influence of the popula- Aptasia pallida. Mar Biol 98:253-262 tion density of zooxanthellae and supply of ammonium on Crossland CJ (1980) Release of photosynthetically-derived the biomass and metabolic characteristics of the reef organic carbon from a , Acropora cf. corals Seriatopora hystrix and Stylophora pistillata. Mar acuminata. In: Schwemmler W, Schenk HEA (eds) Ecol Prog Ser 57:173-186 Endosymbiosis and cell biology. Walter deGruyter, Berlin, Hofrnann DK, Fitt WK, Fleck J (1996) Checkpoints in the life- p 163-172 cycle of Cassiopea spp.: control of metagenesis and meta- Crossland CJ, Barnes DJ, Borowitzka MA (1980b) Diurnal morphosis in a tropical jellyfish. Int J Dev Biol40:331-338 lipid and mucus production in the staghorn coral Acropora Hofmann DK, Kremer BP (1981) Carbon metabolism and stro- acuminata. Mar Biol 60:81-90 bilation in Cassiopea androrneda (Cnidaria: Scyphozoa): Crossland CJ, Barnes DJ, Cox T,Devereux M (1980a) Com- significance of endosymbiotic dinoflagellates. Mar Biol.65: partrn.entation and turnover of organic carbon in the 25-33 staghorn coral Acropora formosa. Mar Biol59: 181- 187 Jassby AD, Platt T (1976) Mathematical formulation of the Davies PS (1984) The role of zooxanthellae in the nutritional relationship between photosynthesis and light for phyto- energy requirements of Pocillopora eydouxi. Coral Reefs 2: plankton. Limnol Oceanogr 213540-547 181-186 Jeffrey SW, Humphrey GF (1975) New spectrophotometric Davies PS (1992) Endosymbiosis in marine cnidarians. In: equations for determining chlorophylls a, b, cl, and c2 in John DM, Hawkins SJ, Price JH (eds) Plant-animal inter- higher plants, algae, and natural phytoplankton. Biochem actions in the marine benthos. Clarendon Press, Oxford, Physiol Pflanz 167:191-194 p 511-540 Kremer P, Costello J, Kremer J, Canino M (1990) Significance DeFreese DE, Clark KB (1991) Transepiderrnal uptake of dls- of photosynthetic to the carbon budget solved free amino acids from seawater by three ascoglos- of the scyphomedusa Linuche unguiculata. Limnol San opisthobranchs. J Molluscan Stud 57:65-74 Oceanogr 35(3):609-624 Drew EA (1972) The biology and physiology of alga-inverte- Lesser MP (1989) Photobiology of natural populations of zoo- brate symbioses I. Carbon fixation in Cassiopea sp. at xanthellae from the sea anemone Ajptasia palljda: assess- Verde & McCloskey: Photobic~logy of the mangrove jellyfish 161

ment of the host's role in protection against ultraviolet tebrates. J Exp Zool 265:410-421 radiation. Cytometry 10:653-658 Prezelin BB (1987) Photosynthetic physiology of dinoflagel- Lesser MP, Shick JM (1989a) Effects of irradiance and ultra- lates. In: Taylor FJR (ed) The biology of dinoflagellates. violet rad~ationon photoadaptation in the zooxanthellae of Blackwell, Oxford, p 174-223 Aiptasia pallida: primary production, photoinhibition, and Rahat M, Hofmann DK (1987) Bacterial and algal effects on enzymatic defenses against oxygen toxicity. Mar Biol 102: mctdlnorphosis In the llfe cycle of Cdsslopea andromeda. 243-255 In: Lee JJ, Fredrick JF (eds) Endocytobiology 111. New Lessc,r MP, Shick JM (1989b) Photoadaptation and defenses York Academy of Sciences, New York, 503:449-458 against oxygen tox~cityin zooxanthellae from natural pop- Richardson K,Beardall J, Raven JA (1983) Adaptation of uni- ulation~of symbiotic cnidarians. J Exp Mar Biol Ecol 134: cellular algae to irradiance: an analysis of strategies. New 129-141 Phyt01 93:157-191 Lugo AE, Snedaker SC (1974) The ecology of mangroves Robertson AI (1988) Decomposition of mangrove leaf litter in Annu Rev Ecol Syst 5:39-64 tropical Australia. .l Exp Mar B101 Ecol 116.235-247 Mackey AP, Smail G (1996) The decomposition of mangrove Schlller C, Herndl GJ (1989) Evidence of enhanced microbial litter in a subtropical mangrove forest. Hydrobiologia 332: activity in the interstitial space of branched corals: possi- 93-98 ble implications for coral metabolism. Coral Reefs 7: McAuley PJ, Cook CB (1994) Effects of host feedlng and dis- 179-184 solved ammonium on cell division and nitrogen status of Schlichter D (1982a) Nutritional strategies of cnidarians: the zooxanthellae in the hydroid Myrjonema arnboinense. absorption, translocation and utilization of dissolved nutri- Mar Biol 121:343-348 ents by Heteroxenia fuscescens Am Zool 22:659-669 McCloskey LR, Aamodt LD, Hazclton WD (1985) A computer- Schlichter D (1982b) Epidermal nutrition of the alcyonarian controlled respirometer for monitoring production and Heteroxenla fuscescens (Ehrb.): absorption of dissolved respiration of symbiotic organisms in situ. Proc 5th Int organic material and lost endogenous photosynthates. Coral Reef Congr 6:137-142 Oecologid 53:40-49 McCloskey LR, Cove TG, Verde EA (1996) Symblont expul- Schlichter D (1984) Cnidaria: permeability, epidermal trans- sion from the anemone Anthopleura elegantissima port and related phenomena. In: Bereiter-Hahn J, Mal- (Brandt) (Cnidaria; ). J Exp Mar Biol Ecol 195: totsy AG. Richards KS (eds) Biology of the integument. l. 173-176 Invertebrates. Springer-Verlag, Berlin, p 79-95 McCloskey LR, Muscatine L, Wilkerson FP (1994) Daily pho- Schlichter D, Kremer BP, Svoboda A (1984) Zooxanthellae tosynthesis, respiration, and carbon budgets in a tropical providing assimilatory power for the incorporation of marine jellyfish (Mastigias sp.). Mar Biol 119:13-22 exogenous acetate in Heteroxenia fuscescens (Cnidaria: McDermott AM, Blanquet RS (1991) Glucose and glycerol Alcyonaria) Mar Blol 83:277-286 uptake by isolated zooxanthellae from Cassiopea xam- Schllchter D, Svoboda A, Kremer RP (1983) Functional .------P'...^^h""; .-.-.P ""A hp hest ua~i>pvl r rrrccllulllSlllil Lb3-.-.- ;rl?t~tr~lphyof H~t~roxnniafuscescens (Anthozoa: Alcy- homogenate fractions. Mar Biol 108:129-136 onaria): carbon assimilation and translocation of photosyn- McDuff RE, Chisholm SW (1982) The calculation of in situ thates from symbionts to host Mar Biol 78.29-38 growth rates of phytoplankton populations from fractions Shick JM, Dunlap WC, Chalker BE, Banaszak AT, Rosen- of cells undergoing mitosis: a clarification. Lirnnol zweig TK (1992) Survey of ultraviolet radiation-absorbing Oceanogr 27:783-788 mycosporine-like amino acids in organs of coral reef Muller-Parker G (1987) Seasonal variation in light-shade holothuroids. Mar Ecol Prog Ser 90:139-148 adaptation of natural populations of the symbiotic sea Shick JM, Lesser MP, Dunlap WC, Stocha] WR, Chalker BE, anemone Aiptasia pulchella (Carlgren, 1943) in Hawaii. Wu Won J (1995) Depth-dependent responses to solar J Exp Mar Biol Ecol 112:165-183 ultraviolet radiation and oxidative stress in the zooxan- Muscatlne L (1990) The role of symbjotlc algae in carbon and thellate coral Acropora microphthalma. Mar Biol 122: energy flux in reef corals. In: Dubinskv Z (ed) Coral reefs 41-51 Elsevier Science Publishers BV, Amsterdam, p 75-87 Smith GJ, Hoegh-Guldberg 0 (1987) Variation in the growth Muscatine L, Marian RE (1982) Dissolved inorganic nitrogen rate of zooxanthellae with coral host colony size is not flux In symbiotic and nonsymbiotic m.edusae. Limnol controlled by changes in the duration of cytokinesis. EOS Octanogr 27:910-917 (Trans Am Geophys Un) 68:1724 Muscatine L. McCloskey LR, Marian RE (1981)Estimating the Sokal RR, Rohlf FJ (1995) Biometry, 3rd edn. WH Freeman daily contribution of carbon from zooxanthellae to coral and Company, New York animal respirat~on.Limnol Oceanogr 26:601-611 Stochaj WR, Dunlap WC, Shlck JM (1994) Two new UV- Muscatlne L, Wilkerson FP, McCloskey LR (1986) Regulation absorbing mycosporine-like amino acids from the sea of population density of symbiotic algae in a tropical anemone Anthopleura elegantissima and the effects of marine jellyfish (Mastigias sp.). Mar Ecol Prog Ser 32: zooxanthellae and spectral irradiance on chemical compo- 279-290 sition and content. Mar Biol 118:149-156 Newel1 SY, Fell JW, Statzell-Tallman A, Miller C, Cefalu RC Strathmann RR (1961) Estimating the organic carbon content (1984) Carbon and nitrogen dynamics in decomposing of phytoplankton from cell volume or plasma volume. leaves of three coastal marine vascular plants of the sub- Limnol Oceanogr 12:4 11 -4 18 tropics. Aquat Bot 19:183-192 Szmant AM, Forrester A (1996) Water column and sediment Olesen NJ, Frandsen K, Rlisgard HU (1994) Population nitrogen and phosphorus distribution patterns in the dynamics, growth and energetics of jellyfish Aurelia Florida Keys, USA. Coral Reefs 1521-41 aurita in a shallow fjord. Mar Ecol Prog Ser 105:9-18 Trench RK (1993) Microalgal-invertebrate symbioses: a Parsons TR, Maita Y, Lalli CM (1984) A manual of chemical review. Endocytobiosis Cell Res 9:135-175 and biological methods for searvater analysis. Pergamon Trench RK, Blank RJ (1987) Symbiodinium microadrjaticum Press Ltd, Oxford, p 101-104 Freudenthal, S. goreauii, sp. nov., S. kawagutii, sp. no\.., Preston RL (1993) Transport of amino acids by marme inver- and S. pilosum, sp nov.: gymnodinioid dinoflagellate sym- 162 Mar Ecol Prog Ser 168: 147-162, 1998

bionts of marine invertebrates. J Phycol23:469-481 zoan Cassiopea xamachana. MSc thesis, University of Vaulot D (1992) Estimate of phytoplankton division rates by Georgia, Athens the mitotic index method: the f,,, approach revisited. Wilkerson FP, Kremer P (1992) DIN, DON, AND PO4 flux by a Limnol Oceanogr 37(3):644-649 medusa with algal symbionts. Mar Ecol Prog Ser 90:237-250 Verde EA, McCloskey LR (1996) Photosynthesis and respira- Wilkerson FP, Muller-Parker G, Muscatine L (1983) Temporal tion of two species of algal symbionts in the anemone patterns of cell division in natural populations of endosym- Anthopleura elegantissima (Brandt) (Cnidaria; Anthozoa). biotic algae. Limnol Oceanogr 28: 1009-101 4 J Exp Mar Biol Ecol 195:18?-202 Zar JH (1996) Biostatistical analysis, 3rd edn. Prentice Hall. Vodenichar JS (1995) Ecological physiology of the scypho- Englewood Cliffs

Editorial responsibility: Otto finne (Editor), Submitted. July 8, 1997; Accepted: March 3, 1998 OldendorfILuhe, Germany Proofs received from a uthor(s): June 12, 1998