Tintinnids ( Protozoa : Ciliophora ) of the Büyükçekmece Bay in the Sea of Marmara * NESLI ·

Ciliates frequently dominate the microzooplankton and have a key position in planktonic food webs as they can respond quickly to phytoplankton pulses composed chiefly of nanoplankton (Capriulo and Ninivaggi, 1982). Tintinnids constitute one major component of marine planktonic ciliates and many species have an apparent cosmopolitan distribution in the seas and oceans (Marshall, 1969). SCI. MAR., 68 (1): 33-44 SCIENTIA MARINA 2004

The Büyükçekmece Bay is located in the northeast of the Sea of Marmara (Fig. 1).It remained connected to Büyükçekmece Lake until 1985, when the connection was blocked by a barrier (11.4 m in height) in order to meet the need for fresh water in Istanbul (Meriç, 1986(Meriç, , 1992)).Since then, the Büyükçekmece Dam Lake has had no effect on the dynamics of the bay because of lack of a water current from the lake to the bay (Meriç, 1992).No previous published studies exist on the tintinnids in the Sea of Marmara, a two-layer water body in which the surface water has characteristics of the Black Sea whereas the deep water has those of the Mediterranean Sea.A study by Sorokin et al. (1995) involving zooplankton reported that tintinnids were rare in the Sea of Marmara and mentioned the abundance of ciliates.There are few data on phytoplankton and their ecological features in the Sea of Marmara (Aubert et al., 1990;Uysal, 1996;Uysal and Ünsal, 1996;Balkıs, 2000Balkıs, , 2003)).Of these, Balkıs (2003) presented data on seasonal variability and abundance of phytoplankton in Büyükçekmece Bay.
The aim of this study is to report on the biodiversity of tintinnids in surface waters in the Sea of Marmara and determine whether the occurrence of tintinnids is correlated to phytoplankton and selected hydrographical factors.

MATERIAL AND METHODS
This research was carried out in Büyükçekmece Bay.Tintinnids and phytoplankton samples for species identification were collected with horizontal tows from the subsurface (0.5 m) with a 55 µm plankton net at five stations (Fig. 1) at monthly intervals from April 1998 to March 1999 and fixed in a 4% neutral formaldehyde solution.The 55-µm net possibly underestimates the abundance of smaller tintinnids due to reduced retention.The species composition sampled with the plankton net should consequently be viewed as size-biased.Identification of smaller species was carried out using a 3 l water sampler at a depth of 0.5 m.Observations of the samples were made through the use of inverted phase contrast microscope equipped with a microphotosystem at a magnification of 400 X.For physical-chemical and quantitative analyses of tintinnid abundance, a 3 l water sampler with thermometer was used at the same depth.These samples were preserved in acidified Lugol's iodine fixative (Throndsen, 1978).Fifty ml subsamples from 3 l water sampler were concentrated by settling in special chambers for 24 h prior to analysis following the method of Utermöhl (Hasle, 1978)  species was recorded.Phytoplankton samples were counted in a Sedgwick-Rafter cell using an inverted phase contrast microscope.Small forms of doubtful taxonomic classification were not added to the list and not counted (Table 1).
At each sampling date measurements of salinity (psu), temperature (°C) and dissolved oxygen (mg l -1 ) were performed (Table 2).The Mohr-Knudsen method (Ivanoff, 1972) was used to measure salinity values, and the Winkler method (Winkler, 1888) to measure dissolved oxygen (DO) values.The abundance of tintinnids and phytoplankton and physico-chemical parameters of the five stations, where the study was carried out, are similar to one another and only the means for all stations are reported.Since the coefficient of variance (V) calculated for five stations for each month was <10% for temperature (0%-5.8%)and salinity (0.3%-1.3%), standard deviations (SD) were not given in the tables and figures.However, V (3.7%-19%) calculated for dissolved oxygen was <10%, with the exception of October-December, so the standard deviation for DO is given in Figure 2.
Spearman rank order correlation was used to correlate abundance of tintinnids with abundance of other phytoplanktonic organisms and hydrographical parameters.Moreover, Nonmetric Multi-Dimensional Scaling (MDS) analysis was performed to estimate relationships between the tintinnid community and hydrographic data.

Tintinnid composition
A total of 14 tintinnid species belonging to 9 genera and 5 families were identified (Table 1).Most of the tintinnids observed belong to the genera Favella and Eutintinnus.The latter was numerically the best represented genus.Total abundance of the tintinnid community varied greatly (Fig. 3).The maximum value for the period of this study was 1.2x10 3 ind.l - 1 in November.The lowest densities were observed in February and March, 0.1x10 3 and 1.6x10 2 ind.l -1 respectively.In January and December no tintinnids were observed.Eutintinnus fraknoi was the most abundant species, with a mean abundance of 0.4x10 3 ind.l -1 .Tintinnid species were observed for ten months within a period of one year.Especially in October and November, it was found that tintinnids increased while the phytoplankton decreased.The abundance of tintinnids was 8x10 2 ind.l -1 in October and 1.2x10 3 ind.l -1 in November.A. amphora, E. apertus, E. fraknoi, E. lusus-undae, F. serrata, H. subulata, M. jörgensenii and S. steenstrupii were the most abundant species.The highest species number was recorded in July (7 species) and November (6 species).The abundance of tintinnids was 7.6x10 2 ind.l -1 in July.The species found during the sampling of this month were C. schabi, E. apertus, E. fraknoi, E. lusus-undae, F. campanula, F. ehrenbergi and T. radix.In February and March only one tintinnid species (F.ehrenbergi) was found (Table 1).
The abundance of tintinnids in Büyükçekmece Bay appears to be negatively correlated to the abundance of total phytoplankton (r s = -0.57,p= 0.05) and diatoms (r s = -0.65,p<0.05) and positively correlated to temperature (r s = 0.61, p<0.05).Furthermore, phytoplankton abundance is positively correlated to dissolved oxygen (r s = 0.64, p<0.05).The other parameters did not appear to play any role in the dynamics of the plankton community of Büyükçekmece Bay.In particular, F. serrata was more affected by temperature and E. lusus-undae by salinity compared to other species (Fig. 4).

DISCUSSION
There are 90 ciliate species known to exist in all the seas of Turkey (Koray et al., 1999).Öztürk (1999) reported 17 whereas Türkog ˘lu and Koray (2000) reported 18 tintinnid species found in the Turkish territorial waters of the Black Sea.Off the coasts of Ukraine, Romania, Bulgaria and Georgia 27, 15, 23 and 9 tintinnid species were found respectively (Petranu, 1997;Zaitsev and Alexandrov, 1998;Konsulov, 1998;Komakhidze and Mazmanidi, 1998).This study reports on 14 tintinnid species found in the Sea of Marmara.All the species found in Büyükçekmece Bay are known to occur in the Aegean and the Mediterranean Seas, while only Coxliella annulata, Favella campanula, F. ehrenbergi, F. serrata, Helicostomella subulata and Tintinnopsis radix are present in the Black Sea.If the sampling had been carried out using a smaller mesh size, more tintinnid species might have been detected.
In this study of the surface waters of the Büyükçekmece Bay the highest numbers of tintinnid species were found in July and November and the lowest number in February and March.In January and December no tintinnids were observed.A nega-tive correlation was observed between tintinnids and the recorded phytoplankton species.In particular, tintinnids increased in both individual number and in species during October and November, when there was a decrease in phytoplankton.In March, in contrast to the increases in diatom abundance, the tintinnid abundance decreased.This may be explained by the general inability of ciliates to feed on colonial diatoms and large dinoflagellates (Hansen, 1991a).Ciliates mainly feed on nanosized prey, preferably nanoflagellates (Burkill et al., 1987;Dolan and Coats, 1990;Paranjape, 1990;Sherr and Sherr, 1994).It is possible that the nanoflagellates were abundant when the large dinoflagellates and diatoms were not, which would explain the negative correlation between tintinnids and the recorded phytoplankton species.Since nanoflagellate abundance was not measured in this study, the role of nanoflagellates remains unknown.However, Aubert et al. (1990) found that nanoflagellates were common in the Sea of Marmara in July (1.3x10 6 ind.l -1 ).This value is concordant with the peak shown by tintinnids in the summer period.Nevertheless, Aubert et al. (1990) did not mention nanoflagellate abundance in November, when tintinnids appear to reach a maximum in this study.
Apart from nanoflagellates, there are several other causes for the negative correlations between tintinnid abundance and phytoplankton abundance.Also, heterotrophic and mixotrophic dinoflagellates are often numerous in marine plankton and are considered important consumers of both phytoplankton and bacteria (Hansen, 1991a;Bockstahler and Coats, 1993a), and they can consume ciliates (Bockstahler and Coats, 1993b).Mixotrophy appears to be widespread among prymnesiophytes and many dinoflagellates (Hansen and Nielsen, 1997;Hansen, 1998).G. sanguineum is one of several species of large mixotrophic dinoflagellates and a predator of ciliates (Bockstahler and Coats, 1993b).In this study, the highest cell number of G. sanguineum was found in May (1.3x10 3 ind.l -1 ).Also, Dinophysis hastata and Phalacroma rotundatum can ingest ciliates (Hansen, 1991b), but these species did not reach great numbers during the sampling period.Other mixotrophic dinoflagellates such as Ceratium and Dinophysis, and heterotrophic ones such as Diplopsalis, Gymnodinium, Noctiluca and Protoperidinium were found throughout the year.However, the abundance was generally low and it is thus not likely that there was any major grazing pressure from dinoflagellates on ciliates.It is more likely that other preda-tors, i.e. mesozooplankton, were more important consumers of the tintinnids (Turner and Anderson, 1983;Turner et al., 1998;Coats and Revalente, 1999;Levinsen and Nielsen, 2002).Since, mesozooplankton was not measured in this study, we do not know the extent of this predation, but it would be interesting to study it in the future since it could explain some of the seasonal patterns of the tintinnids that were found.Only Uysal (1996) reported individuals of different zooplankton groups formed by copepods, siphonophores, chaetognaths, polychaete larvae, cladocerans and appendicularians in the Sea of Marmara.The percentage distribution of zooplankton groups revealed that the predominance of copepods persists throughout the year in the region, and the highest recorded zooplankton level for the upper layer was 125400 ind./m 3 in September 1985, and the lowest was 3980 ind./m 3 .
Other factors that can cause the negative correlation between phytoplankton and ciliates include the possibility that some phytoplankton may produce chemical defence compounds.The best-known are toxin-producing dinoflagellates, which may have a negative impact on tintinnids (Hansen et al., 1992).A number of marine dinoflagellates have been known to produce nonprotein toxins, and these dinoflagellates are capable of forming red tides that inhibit zooplankton grazing (Hansen, 1989) late spring/summer, peaks for phytoplankton abundances and production have been recorded in many Mediterranean coastal regions.Red tides are also more frequent at these times of the year (Zingone et al., 1990).In this study, none of the phytoplanktonic species exceeded one million cells per litre of surface water.At no time during this study was any colouring of the surface water detected.Despite the presence of certain dinoflagellate species (Ceratium furca, Dinophysis acuta, Heterocapsa triquetra, Lingulodinium polyedrum, Noctiluca scintillans, Phalacroma rotundatum, Prorocentrum micans, P. triestinum, Scrippsiella trochoidea) responsible for red tides and other noxious algal blooms in other geographic areas (Koray et al., 1992;Hallegraeff, 1993;Smalley and Coats, 2002), red tides were not recorded during the sampling period of this study.Throughout the year, the tintinnid species were found in the range of 7. 3-23.5°C, 19.7-23.3 psu and 7.13-11.95mg l -1 .These values are characteristic for this area (Ünlüata et al., 1990;Bes ¸iktepe et al., 1995) and the chemical oceanography of the Sea of Marmara is significantly influenced by the biochemistry of the Black Sea and the Aegean Sea.It connects to the Black Sea through the Bosphorus in the NE and to the Aegean Sea via the Dardanelles in the SW.The basin is occupied by two distinctly different water masses throughout the year: the brackish waters (22-26 psu) of the Black Sea origin, forming a relatively thin surface layer (10-15 m thick) with a mean residence time of about 4-5 months, and the subhalocline waters of Mediterranean origin (38.5-38.6 psu) separated from the former by a sharp interface (pycnocline) about 10-20 m thick.Because of the large volume of water inflow from the adjacent Black Sea (about 600 km 3 ) into the relatively small upper layer volume (about 225 km 3 ) of the Sea of Marmara, the upper layer ecosystem of the latter has been influenced to a large extent.(Ünlüata et al., 1990;Tug ˘rul and Polat, 1995).In particular, at depths of 0.5-20 m, the Sea of Marmara is known to be affected by the brackish water coming from the Black Sea via the Bosphorus (Yüce and Türker, 1991).The abundance of tintinnids has been effected by different water masses in the area.It was observed that the maximum abundance of tintinnids was found when salinity was high and temperature low.It is known that there are limited vertical exchanges between water masses due to thermocline and halocline layers, particularly during spring and summer, and the water on the surface does not usually sink down to the bottom.In autumn and winter, winds cause the water to become rough, the stratification is broken up, and the water from the bottom comes up to the surface (Balkıs, 2003).Such a phenomenon is important for the transport of tintinnids to the upper strata and may explain the maximum abundance of tintinnids in autumn.Moreover, on the surface, the water is usually over-saturated due to the exchanges with the atmosphere.Mixed water during the period of October to December may be the reason for the differences between the stations, especially in the O 2 values (V>10%).
This study is the first to report on the composition and abundance of tintinnid species in the Büyükçekmece Bay, and the photographs of species have been illustrated (Figs. 5,6).The abundance of tintinnids was negatively correlated with that of large phytoplankton species, which is probably due to their inability to consume these large prey.It is more likely that the tintinnids prey on nanoflagellates, but these were not included in this study and need to be explored in the future.
FIG. 4. -Multi-Dimensional Scaling (MDS) plot in two dimensions for the tintinnid community in Büyükçekmece Bay with relation to temperature, salinity and oxygen.