Members of AUSCORE met for the 3^rd Annual AUSCORE Workshop in February 2000. The Workshop was hosted by the Research School of Earth Sciences, Australian National University, Canberra. The group discussed past and future research plans. The following abstracts give an indication of the range of research into coral records covered by the group. The 4^th Annual AUSCORE Workshop is planned to be held in Brisbane in February 2001. For more information about our research activities contact individual authors – see also profiles of our members.

In-shore corals of the Great Barrier Reef are particularly exposed to extreme environmental changes related to terrestrial inputs. Increased water turbidity from river plumes carrying fine suspended sediments or from resuspension of bottom sediments (e.g. along the track of tropical cyclones), transient low salinity episodes (heavy rainfall and freshwater plumes) or enhanced warming of coastal waters such as during Feb ’98, are some examples of the potentially stressful events that these corals experience during their lifetime.

Two major markers of terrestrial inputs with contrasted geochemical behavior: Ba and Mn have been measured using laser ablation ICP-MS in the skeleton of a Porites coral from Pandora Reef, located near Townsville at ~15 km from shore and ~150 km to the North of the mouth of the Burdekin River. This record covers the recent period 1998-89 including the ’91 and ’98 major flood events in the wake of tropical cyclones "Joy" and "Sid", respectively. The accuracy and robustness of a range of geochemical tracers (Sr/Ca, B/Ca, U/Ca, Mg/Ca. Mn/Ca and Ba/Ca) are assessed against well-documented SSTs, rainfall, river discharge and salinity data. Interannual and seasonal changes of the coral extension rate are estimated from the timescale ascribed to the Sr/Ca record by fitting to instrumental surface water temperatures. For this Pandora coral, Sr/Ca and B/Ca appear as the best proxy thermometers. Ba/Ca is distinct from the other ratios as it does not show seasonal variations but well-defined peaks associated with the fresh water plumes of the Burdekin River deflected northward along the shore by currents. The timing and intensity of the coral Ba peaks are compared with the residence time of river plumes in the vicinity of Pandora Reef, river flow and minimum salinity figures provided by a recent study at AIMS.

Holes and indentations pushed into laboratory grade calcium carbonate powder showed a yellow luminescence indistinguishable from luminescence seen in coral skeletal slices.  Observations and measurements of luminescence in hole pushed into laboratory grade calcium carbonate powder reproduced all the features of luminescence in coral skeletal slices. Long wavelength ultraviolet (UV) light from fluorescent tubes and other sources used to display coral luminescent banding and lines contains significant amounts of violet and blue visible light. Luminescence in coral skeletons is excited by wavelengths from UV through to green. Light returning from holes and indentations in coral skeletons will have undergone more reflections than light returning from cut skeletal surface. Each reflection increases the probability of absorption of UV, violet and blue light and its re-emission at longer wavelengths (luminescence). Thus light returning from holes and indentations in coral skeletal slices will contain relatively less violet and blue visible light than light returning from sawn surfaces of the slice. Thus, light returned from surfaces appears blue while light returned from holes and indentations appears yellow.

The yellow luminescence seen in slices of coral skeletons and the blue luminescence measured in such slices are properties of mineral calcium carbonate. In corals, enhanced luminescence is associated with regions with larger numbers of holes and indentations. The luminescent lines associated with monsoonal river flows in corals from the Great Barrier Reef are narrow regions of lower density skeleton i.e., regions with greater amounts of holes and indentations. These narrow, low-density regions presumably result because significantly lower salinities reduce coral calcification without concomitant reduction in skeletal extension. Offshore corals, not subject to regular, periodically lowered salinities show luminescent banding in which higher luminescence is associated with the lower density portion of the annual skeletal density banding pattern.

Records associated with skeletal stable isotopes and trace skeletal inclusions are usually obtained from the surface of a coral skeletal slice. Luminescence is a representation of skeletal density at the surface of a skeletal slice rather than density averaged over the thickness of a slice, as is obtained with conventional measurement techniques such as X-radiography and gamma densitometry. Thus, luminescence offers a means to directly align these different records.

We report measurements of the intra-annual variation in extension rate in colonies of Porites. These measurements were made by comparing the annual temperature signal recorded in coral skeletons as d¹⁸O ratios with the actual temperature signals.

Although the theoretical relationship between temperature and d¹⁸O ratios in skeletal carbonates is not linear, the relationship is generally considered linear over the temperature ranges experienced by corals. Changes in the constants used to describe the linear relationship do not affect results given here. The annual seawater temperature signal was assumed to be a sine curve as a function of time. This was converted to a temperature signal as a function of distance along a coral slice using the relationship distance = time + constant * sin(2*time). Thus, when the constant = 0 there were no intra-annual variations in extension rate (distance µ time). Where constant = 1 the coral momentarily stopped growing in midwinter. This modified sine curve for temperature was fitted to the isotopic data and the constant varied until the best least squares fit was obtained. In the tropics, the annual seawater temperature signal is not a sine curve. Consequently, we modified the temperature signal using the relationship above but with distance replaced by temperature until we obtained the best least squares fit between actual temperatures and our modified sine curve temperatures. This provided two constants, 1 for fits of modified temperature to isotope data and the other for fits of actual temperatures to modified temperatures. The difference between these two constants is a measure of intra-annual variation in extension.

Data for isotopic ratios was provided by Dr Mike Gagan and came from a colony of Porites lutea from Pandora Reef in the central Great Barrier Reef and represented the period 1977-84 (see Gagan et al., 1994). The theoretical relation between distance along the coral growth axis and the isotopic ratio can be calculated and the important parameter is the intra-annual change in extension rate (Barnes et. al, 1995). Results given here are not dependent upon the amount of thickening applied to vertical skeletal elements. The same results would be obtained if all skeletal growth occurred as extension. The results for three cases are provided (Fig. 1). Fits between isotope rations and the modified sine curve temperature data were obtained for the entire data set and also for selected regions. Figure 2 gives the best fit for constant intra-annual extension rate and the best fit obtained by varying extension rate. The fitted period corresponds to early autumn to late spring. The period over summer was excluded to avoid rainfall effects on isotope ratios (eg, Fig 3 see Gagan et. al., 1994). Figure 2 shows the better fit is obtained with a variable extension rate. The entire data set also shows a significantly better fit for a variable extension rate (Fig. 3).

Fits made as in Figure 2 indicate about a 3-fold variation in extension rate (max extension rate / minimum extension rate ~3). The fit for the entire data set (Fig. 3) gave a value of 2. A lower value for the fit for the entire data set was expected because of the two years in which isotopic ratios peaked well above expected values, which correspond to high rainfall years (Gagan et al., 1994). When allowance was made for the non-sinusoidal temperature curve (i.e. by subtracting the constants; see above), fits made as in Figure 2 returned a 2.5-fold variation in extension rate. It appears that this colony of P. lutea averaged growth that was 2.5 times faster in summer than winter over the period 1977-84.

References

1. Barnes, D.J., Taylor, R.B., & Lough, J.M.,1995. On the inclusion of trace materials into massive coral skeletons. Part II: distortions in skeletal records of annual climate cycles due to growth processes. J. Exp. Mar. Biol. Ecol., Vol.194 pp.251-275.2. 3. Gagan, M.K., Chivas, A.R. and Isdale, P.J., 1994. High resolution isotopic records from corals using ocean temperature and mass spawning chronometers. Earth and Planetary Sci. Let., 121 , pp. 549-558.4.

Figure Captions

1. Theoretical relationship between the isotopic ratio and distance along the coral. a) constant extension, b) max. ext. rate/min. ext. rate = 2, c) coral just stops growing in mid winter.2. 3. Best theoretical fit to part of the data. a) constant extension, b) extension rate ( variable V) adjusted to give best fit 4. 5. Best theoretical fit for all the data. a) constant extension, b) extension rate adjusted to give best fit.

In 1989 open pit gold mining commenced on the island of Misima in Papua New Guinea. Open pit mining by its nature causes a significant increase in sedimentation, both natural and mining-induced. This increased sedimentation affected the nearby fringing coral reef to varying degrees, causing coral mortality (complete suffocation) in some areas. This sediment consists of soft mine waste which is made up of quartz feldspar, greenstone and schist. These rocks have distinct chemical constituents (rare earth elements [REE], zinc and lead etc.), that are entering the near-shore environment in considerably higher than normal concentrations. In this study we evaluate whether Porites corals can be utilized as a tool for recording environmental input of trace elements in near-shore environments.

Density, extension and calcification was examined by Barnes and Lough (1999) for these Porites coral colonies, they suggest that high sedimentation may not be recorded by these growth characteristics. However, they did note a positive correlation between decreasing coral tissue layer thickness and proximity to the highest sedimentation areas. This is consistent with corals under stress (Barnes and Lough, 1992). Using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) we analyzed four colonies (2 from high sedimentation, 2 control sites) for uranium, cerium (REE), zinc, manganese, lead and barium. The two "severely" affected corals show low steady "background" levels prior to the commencement of mining. After 1989 they show dramatic increases of cerium, zinc, manganese and to some degree lead. The control sites, which are distal from the mining operations, do not show similar increases in these elements after mining. In conclusion, trace element studies of Misima Island corals clearly record the dramatic changes in the environmental conditions at this site and provide a basis to test the subtle anthropogenic influences on corals in the GBR.

References

Barnes, DJ, Lough, JM (1992) Systematic variations in the depth of skeleton occupied by coral tissue in massive colonies of Porites from the Great Barrier Reef. J Exp Mar Biol Ecol 139:113-128.

Barnes, DJ, Lough, JM (1999) Porites growth characteristics in a changed environment: Misima Island, Papua New Guinea. Coral Reefs 18:213-218.

POSTER

The rapid high-resolution analytical technique of Laser Ablation ICP-MS is used to measure boron, magnesium, strontium, barium, and uranium abundance in Porites corals and coralline sponges. Cores from Porites sp. coral colonies were collected from inshore, mid-shelf and outer reef localities (central GBR) to test the robustness of these elements as environmental proxies. The inshore reefs selected for this study are heavily influenced by river runoff whereas the mid-reef and outer-reef locations are not. This is clearly recorded by Ba/Ca, which provides an excellent proxy for river run-off. Time series analyses of Sr/Ca, U/Ca, B/Ca and Mg/Ca are compared to in-situ sea surface temperature (SST) and/or IGOSS NMC weekly SST to provide calibrations for these elements. Previous workers have noted differences in the calibration of Sr/Ca vs. SST; this LA-ICP-MS data-set shows a slight variation between different corals, although the TIMS Sr/Ca is highly reproducible. This suggest small scale intra-coral variability. Both the U/Ca and Mg/Ca have calibrations within error for mid-shelf and outer reef corals but the calibrations differ for the inshore corals. The B/Ca appears to be more robust in terms of its temperature calibration.

Recently, coralline sponges have been proposed as a new source of tropical paleoclimatic information. Profiles of d¹³C in coralline sponges have documented (better and more accurately than corals) the atmospheric increase of d¹³C associated with increased fossil fuel consumption. Sponges were collected from 2 locations on the outer GBR, Ribbon Reef #10 and Myrmidon Reef. Preliminary results suggest these sponges will provide meaningful proxy information about past conditions in the same manner as corals. Due to their very slow growth rates (~0.2 mm/yr) sponges are better suited to recording and providing long-term environmental information. When samples are compared at longer-term (annual to several years), patterns appear which are consistent between Sr/Ca, Mg/Ca, B/Ca and Ba/Ca. U/Ca appears to vary both in and out of phase with respect to the other three elements. The B, Mg and Ba concentrations are 2-5 times lower than in corals with concentrations of ~20 ppm, ~200 ppm and ~4 ppm, respectively. The Sr and U concentration are 1-2.5 times higher than in corals with concentrations of ~9000 ppm and ~8 ppm, respectively.

Annual extension, density and calcification were measured in 245 similar-sized colonies of Porites from similar locations on 29 reefs from across the length and breadth of the Great Barrier Reef (GBR). Average growth characteristics were obtained for an eight-year period common to all colonies. Extension rate and calcification rate decreased from north to south along the GBR (latitudinal range ~9^o) and were significantly and directly related to annual average sea surface temperature (SST; range 25-27^oC). For each 1^oC rise in SST, average annual calcification increased by 0.39 g.cm^-2.yr^-1 and average annual extension increased by 3.1 mm/yr. Density was inversely correlated with extension rate and increased with distance from shore. Data for massive Porites colonies from the GBR were extended through 20^o of latitude and an average annual SST range of 23-29^oC using published data for the Hawaiian Archipelgo (Grigg, 1981) and Phuket, Thailand (Scoffin et al., 1992). The response of calcification rate to SST remained linear. Variation in annual average SST accounted for 84% of the variance. For each 1^oC rise in SST, average annual calcification increased by 0.33 g.cm^-2..yr^-1 and average annual extension by 3.1 mm.yr^-1. Annual minimum SSTs were more closely related to calcification and extension rates than annual maximum SSTs. Inter-annual variations of calcification in Porites were previously reported (Lough & Barnes, 1998) to be about half as sensitive to variations in SST. These data are shown, however, to closely match the linear fit based on 44 Indo-Pacific sites. Thus, variation in calcification rate of Porites with average annual SST appears to be similar whether examined over space or time. The sensitivity of calcification rate in Porites to SST, combined with observed 20^th century increases in SSTs, suggests that calcification rates may have already significantly increased along the GBR in response to global climate change.

Four Porites sp. corals, with U/Th ages of 7600-8000 years BP, were recovered from the base of a drill hole on the windward margin of Myrmidon Reef. Sr/Ca ratios have been measured for each coral by isotope dilution thermal ionisation mass spectrometry. These ratios have been converted to sea surface temperature (SST) by using the calibration of a modern Porites sp. with a 10 year instrumental record from the same reef (Sr/Ca.10³ = 10.40 - 0.0575T). Two corals show SST’s that are about the same as modern values, but one coral shows SSTs that are 1.5-2.0°C warmer than present and another 1-2°C cooler. Two of the corals show distinct cooling trends which resemble an El Nino signal in modern corals from the region. The difference in SSTs for these chronologically similar corals could be related to some not being sampled along the major growth axis.

An unusual feature of these corals is that they show fluorescent bands, which is normally associated with river runoff events in near-shore corals. The modern Myrmidon corals do not show fluorescence. Also, there is the presence of distinct Ba/Ca peaks, as measured by laser ablation ICP-MS. Again, in near-shore corals these anomalies have been interpreted as runoff events. However, the distance from Myrmidon Reef to the coast of over 100 km precludes any effect of river runoff, even at 8000 years BP when sea level was some 20 m lower than present. This is confirmed by the d¹⁸O signal which, when corrected for sea level, matches the Sr/Ca curve. The fluorescence and Ba/Ca anomalies suggest that upwelling along the edge of the shelf was more prevalent than it is today.

While the Sr/Ca data suggests that SSTs were similar to the present day, the d¹⁸O-derived SSTs, calibrated against an inshore coral, are cooler by 2-3°C. The lower SST range is more consistent with Sr/Ca-derived SSTs in corals of similar age from the Huon Peninsula and Vanuatu. However, it is considered that the Sr/Ca derived SSTs are more reliable because of their calibration with instrumental and modern coral data from the same reef.

The inshore region of the Great Barrier Reef is subject to the combined stresses of terrestrial run-off from flood plumes as well as unusually warm SST’s that can induce coral bleaching. In February-March 1998, as part of a global phenomenon, the inner GBR experienced an episode of large-scale mass coral bleaching that was probably the most intense and widespread yet recorded. More than two thirds of the inshore reefs exhibited high-levels of bleaching with a quarter being subject to extreme (irreversible?) bleaching (Berkelmans and Oliver, Coral Reefs 18, 55-60, 1999). Immediately prior to this event, incursion of low salinity flood plumes occurred in the inshore region of the central GBR as a result of a monsoonal depression. To better quantify the response of Porites coral to both river run-off and possible impacts on coral bleaching, we report here high-resolution laser ablation and mass spectrometric trace element analyses of Porites coral from the inshore Frankland Island and Pandora Reefs.

Trace element proxies for sea surface temperature (Sr/Ca, U/Ca and Mg/Ca) and river run-off (Ba/Ca, Ce,/Ca, Y/Ca) were determined in corals collected 8 months after the mass bleaching episode from the inshore reefs at Frankland Islands and Pandora Reef. All the corals clearly record the high temperature bleaching event, indicating that calcification continued despite SST’s being in the range of 30-31^oC. For 2-4 weeks following the bleaching episode, some growth (extension) of the coral polyps continued, although at a much lower density, indicating a major decrease in the overall rate of calcification, presumably due to the expulsion of symbiotic algae. This latter phase is recorded as a rapid increase in both the Sr/Ca and U/Ca ratios. During the bleaching event, corals also show anonymously high Mg/Ca, another possible fingerprint for bleaching. The longer-term response to the bleaching is variable. One coral shows relatively minor effects, corresponding to an ~1 month hiatus in calcification. Another coral still containing tissue, remains dormant after 8 months, showing no signs of renewed calcification. Finally there are many dead Porites where macro-algae has resorbed parts of the former coral tissue.

In early January 1998, the flood plumes that entered the inner GBR from the Russell-Mulgrave and Burdekin Rivers impacted corals at Frankland Islands and Pandora Reefs respectively. The timing and approximate magnitude of the flood peaks is faithfully recorded in the Ba/Ca ratios in corals from these localities. In river systems Ba is strongly particle reactive, only undergoing desorption as it enters the estuarine zone. Thereafter, Ba probably undergoes approximately conservative mixing, although this remains to be established for the GBR over a wide salinity range. The self-consistent records from the Frankland Islands and Pandora Reefs appear to substantiate this inference and implies that Ba/Ca ratios in corals may provide a proxy for riverine suspended sediment loads being delivered to the inshore GBR (Sinclair and McCulloch, 2000). Bleaching in the inshore GBR may have been exacerbated by the formation of a relatively stable low salinity ‘lid’ that promoted rapid warming of shallow waters. Studies are using long-lived (400-500 year old) Porites corals may thus provide a better understanding of possible links between flood plumes and severe episodes of coral bleaching.

At present, the conventional way of measuring coral UV fluorescence in Australia involves the use of a fluorometer (Fluorac) at the Australian Institute of Marine Science. As an alternative to Fluorac digital image analysis (DIA) of coral UV fluorescence has been developed. With the DIA technique, the grey level of a digital image of coral fluorescence is an indirect measure of fluorescence. DIA is a cheap alternative to measuring UV fluorescence on a fluorometer and, if set up correctly, will facilitate fast acquisition of fluorescence data.

The digital image of the coral UV fluorescence is obtained by photographing the coral using conventional methods, then scanning the image into a computer. Initially markers, used to define the image scale, are placed on an ultrasonically cleaned 7mm thick coral slice. The coral slice is illuminated under two UV tubes and black and white photographs are taken. A yellow filter is used to cut out the blue part of the spectrum. Focus shifting, aperture, exposure times and the placement of the coral in the photograph are set to increase image clarity and to minimise any potential lens edge distortion effects. A standard piece of coral is also placed in the same position in each photograph so that the photos can be calibrated. Finally, coral photographs are scanned into a computer with scanning parameters set to minimise possible changes to the image brightness and contrast during the scanning process.

Coral image grey levels are measured using NIH Image v 1.61 software available free from the Internet at http://rsb.info.nih.gov/nih-image/download.html . Changes in grey level on the image represent the changes in intensity of the fluorescent banding. These changes are measured along a user-defined sample path and the image grey levels are calibrated and the scale set in the NIH Image program.

To date, visual comparison of the image grey level with coral fluorescence is favourable. There are however, some changes in image contrast induced by the scanning software and testing is under way to remove these changes. The final stage in the development of this procedure will be to compare results from DIA with those from Fluorac, (ensuring that the same data processing methods are used), and comparing these datasets with river discharge data. Corals from Pandora Reef, Queensland and Blup Blup Island, Papua New Guinea will be used for the comparitive study. Reproducability of coral UV fluorescence will also be investigated, along with changes in Sepik River discharge, using two corals from the same site at Muschu Island, Papua New Guinea.

POSTER

Massive coral such as the genus Porites are increasingly used as climate proxies in the tropics where instrumental data is insufficient or lacking. More work is underway to extend the methods used such as measuring ?d¹³C, d¹⁸O and Sr/Ca ratios. The objective of this paper is to monitor coral response to changes in environmental parameters in Ningaloo Reef and to show the reproducibility of results from 5-year increments over 150 years before pursuing high resolution work for selected time spans. The importance of a careful petrographic and mineralogical investigation of coral cores prior to geochemical analysis is stressed, because small post-depositional alteration of the coral skeleton appear to cause significant shifts in the geochemical signal and mask environmental signals. We also show that a multi-proxy approach can considerably increase confidence in the interpretation of geochemical data from massive coral.

Two long coral cores of Porites lutea from Ningaloo Reef, covering time periods of 168 and 140 years have been investigated. Stable isotopes of carbon (d¹³C) and oxygen (d¹⁸O) have been measured on 5-year increments of these cores to demonstrate the reproducibility of 150-year trends and to determine the magnitude and reproducibility of interdecadal variability for both coral colonies. Sr/Ca ratios have been measured for every second 5-year increment to allow differentiation between temperature and salinity signals of the d¹⁸O-record.

The ?18O values in both cores suggest very stable sea surface temperatures (SST) over the last 100 years. However, the cores show different magnitudes of the d¹⁸O suggesting that the values are not controlled by SST alone. The differences between the Sr/Ca and d¹⁸O curves, the residual d¹⁸O signals (Dd¹⁸O), suggest the presence of different salinity regimes at the sampling sites. High residuals coincide with high SST in the core taken west of North West Cape, suggesting enhanced evaporation associated with enrichment of ¹⁸O in seawater in a lagoonal enviroment. In the bottom part of the second core taken north of North West Cape, secondary aragonite is present, the geochemical data shows the strongest anomalies, and there is good correlation between d¹⁸O, d¹³C and Sr/Ca ratios. The magnitude of variation in these ratios suggest a strong imprint from a secondary precipitation of aragonite, obscuring any temperature signal.

Modern analytical methods such as laser-ablation ICP-MS allow coral skeletons to be simultaneously analysed for a number of chemical tracers. At the RSES it is common to analyse a suite of elements including B, Mg, Sr, Ba, U, Y, La, Ce, Mn, Zn, and P. Typically B/Ca, Mg/Ca, Sr/Ca, and U/Ca vary seasonally in Porites corals, apparently linked to cycles in SST. These 4 trace elements do not correlate exactly with each other in corals, however. It therefore appears as if other chemical or biological factors can influence each trace element independently, producing subtly different variations on a straight SST profile. As more corals are analysed, patterns are beginning to emerge which may give clues as to what chemical or biological factors influence trace element signals.

As an example, a regular -ve spike occurring during mid summer destroys the usually strong seasonal cycle in Mg/Ca profiles in a number of corals. A similar phenomenon occurs in U/Ca profiles, although the magnitude of the distortion relative to the amplitude of the seasonal signal appears to be less than in the Mg/Ca signal. One interpretation of this observation is that Mg and U respond independently to two environmental parameters. SST variations give rise to a seasonal signal, which is countered by an unknown parameter for which Mg is more sensitive than U.

In this presentation, I will give examples of trace element signals that deviate from a seasonal SST cycle. I will also explore a number of hypothetical models, and discuss how this research might proceed in the future.

Alcyonacean corals are formed by tissues, mesoglea and sclerites. Their colonial, benthic growth is associated with their use of secondary metabolites in the competition for space amongst other problems of life in coral reefs. Other biological attributes, still not fully studied, allow them to occupy areas beyond the distribution boundaries of scleractinian corals. The sclerites make up to 36% of the colony volume fraction. Like other biomineral structures, sclerites are a potential reservoir of enviromental information. Before using them as indicators, we need to answer these questions: How fast are soft coral sclerites formed? Is calcification affected by environmental parameters? What is the ‘life-span’ of soft coral sclerites? Is calcification coupled with cellular growth in soft corals? Experimental work addressing these questions and carried out after the first AUSCORE meeting is summarized here.

1. Not all regions of the soft coral colony calcify at the same rate. The base of the colony has a significantly higher calcification rate than the top of the colony as measured by ⁴⁵Ca labelling of sclerites of Litophyton arboreum. Cellular growth rates follow a similar pattern, however, both measurements have not been performed simultaneously yet and it is not known if both processes are coupled.

2. Calcification is sensitive to light but not in the same fashion as observed in scleractinian corals. Soft corals demineralize in the absence of light as shown by preliminary experiments on Sinularia sp. using the alkalinity anomaly technique.

3. Mg/Ca ratios show significant variations between sclerites obtained in winter and summer times. During winter, the Mg/Ca ratio is lower than during summer. These measurements were made on Sinularia and Dendronephthya collected in the fringing reefs of Keppel Bay, GBR. These variations seem to be associated with light intensity and/or water temperature. Mg/Ca ratios performed on sclerites from Sinularia, Anthelia, Sarcophyton, Litophyton collected through autumn, winter and summer time at the aquarium of the Museum of Oceanography of Monaco did not show significant variation. This aquarium is kept under constant temperature, salinity and light regimes.

4. All sclerites are surrounded by living cells. Most of the sclerites are located in the mesoglea. These observations have been made by TEM on Sarcophyton sp.

A model of calcification in soft corals and their potential as bio-indicators will be discussed.