Feeding and predation impact of chaetognaths in the north Aegean Sea (Strymonikos and Ierissos Gulfs)

Transcription

1 J. Mar. Biol. Ass. U.K. (2005), 85,1525^1532 Printed in the United Kingdom Feeding and predation impact of chaetognaths in the north Aegean Sea (Strymonikos and Ierissos Gulfs) George Kehayias* $,EvangeliaMichaloudi O and Emmanuil Koutrakis P *Department of Environmental and Natural Resources Management, University of Ioannina, Seferi 2, Agrinio, Greece. O Aristotle University of Thessaloniki, Department of Biology, Ichthyology Laboratory, Box 134, Thessaloniki, Greece. P Fisheries Research InstituteöNational Agricultural Research Foundation, Nea Peramos, Kavala, Greece. $ Corresponding author, gkechagi@cc.uoi.gr The feeding of the most important chaetognath species (Sagitta en ata, Sagitta minima, Sagitta setosa and Sagitta serratodentata) found in a grid of 35 stations in the north Aegean Sea (Strymonikos and Ierissos Gulfs) was investigated through gut content analysis during ve sampling periods from June 199 to May. Sagitta en ata and S. minima were the most abundant species in summer ^autumn 199 and in spring, respectively. Copepods were the main food for all chaetognath species in spring, while cladocerans in summer^autumn 199 dominated the diet, especially of S. en ata. In September 199, the juvenile specimens of cladocerans were the most important prey of the immature Stage I specimens of S. en ata, which resulted from the major breeding period in late summer. The estimated impact of chaetognath predation on the copepod and cladoceran communities ranged between 0.02^2.6% and 0.01^1.29% of the copepod and cladoceran standing stock, respectively. Stage I specimens of S. en ata accounted for nearly 90% of the total copepod and cladoceran standing stock consumed per day by the total population of this species in September 199. INTRODUCTION Chaetognaths are considered among the most important secondary consumers in marine ecosystems and although they feed mainly on copepods, occasionally they can use a wide variety of other zooplankters as food (Feigenbaum, 1991). The feeding and predation impact of chaetognaths on their prey have been the subject of only two studies in the Mediterranean: Duro & Saiz (2000) in coastal and pelagic areas of the western basin and Kehayias (2003) in pelagic areas of the eastern basin. Further, there is no information about the diet and the predation impact of chaetognaths in the Aegean Sea. At the central part of the northern Aegean Sea lie Strymonikos and Ierissos Gulfs (40848 N E and N E) covering an area of 630 km 2. These two gulfs are a ected by freshwater input from the Rivers Strymon and Rihios, which are the main sources of nutrients and pollutants (domestic and agricultural) into the sea resulting in high productivity (Stamatis et al., 2001). The aim of this study is to provide new information about the ecological role of chaetognaths as secondary consumers in this productive coastal area of the eastern Mediterranean, by studying their diet and predation impact during ve sampling periods. MATERIALS AND METHODS Field sampling Zooplankton samples were collected in June, July and September 199 and March and May at 35 stations (Koutrakis et al., 2005). For the collection of the samples a Bongo net (mouth area m 2,meshsize500mm) was used (UNESCO, 1968). A owmeter was tted at the mouth of the net for the estimation of the water volume ltered. Oblique hauls were taken at a speed of 10^20 m min 1 from within 5 m of the bottom to the surface. All stations were sampled during the day. The samples were xed immediately after collection and preserved in 4% formalin solution in seawater. Tow duration was in most cases more than 2 min, while the time from the beginning of the tow to the preservation of each sample was greater than 5 min. Laboratory methods All chaetognath specimens were sorted from the samples, identi ed to species level and classi ed into four maturity stages using the system of Kehayias et al. (1999). The specimens apparently containing food were dissected and the items were examined. In order to account for codend feeding, chaetognaths in which the prey was found in the forward 1/3 of the gut were omitted from the counts. Food items were classi ed only to the level of zooplankton groups. The food containing ratio (FCR), expressed as the ratio of chaetognaths containing food to the total chaetognaths examined (Feigenbaum, 1991) was estimated for each species and every station. Ingestion rates (I, prey ingested per chaetognath per day) were calculated using a modi ed version of Bajkov s (1935) equation: I ¼ 2 NPC DT 24 (1)

2 1526 G. Kehayias et al. Feeding and predation impact of chaetognaths Table 1. The percentage (%) of the ontogenetic Stages (I^IV) within each chaetognath species population and abundance (as individuals per m 3 ) of the total chaetognaths and each species during ve sampling periods in the north Aegean Sea. Values are expressed as average standard error. Species Stage June 199 July 199 September 199 March May Sagitta en ata I II III IV Ind per m 3 Total Sagitta minima I II III IV Ind per m 3 Total Sagitta setosa I II III IV Ind per m 3 Total Sagitta serratodentata I ^ ^ ^ II ^ ^ ^ III ^ ^ ^ IV ^ ^ ^ Ind per m 3 Total ^ ^ ^ Total chaetognaths where DT is the digestion time (in hours) and NPC is the average number of prey per chaetognath (Feigenbaum, 1991). The factor two in the equation accounts for the loss of prey due to gut evacuation during sampling and preservation (Baier & Purcell, 199). When only one prey item is found in the gut content, as it was in the present study, NPC value is equal to that of FCR. Thus, the equation can be transformed to: I ¼ 2 FCR 24 (2) DT The digestion time (DT) for all chaetognath species was calculated using the empirical equation: DT ¼ 10.96e 0.086T derived for the Sagitta sp. by Stuart & Verheye (1991). In this equation, DT is the digestion time in hours and T is the temperature in 8C. The mean temperatures used in the digestion rate equation were: 18C in June, 188C in July and 198C in September 199, 128C in March and15 8C in May (Sylaios et al.,2005). The predation pressure (prey consumed per m 3 per day or P m 3 d 1 ) of each chaetognath species for the ve Figure 1. Food containing ratio (FCR) values (means SE) of the total chaetognaths and the individual species found in Strymonikos and Ierissos Gulfs during the ve sampling periods.

3 Table 2. Composition of the gut contents (%) and total number of prey items of the four chaetognath species found in the north Aegean Sea during ve sampling periods. Prey Ontog. stages June 199 July 199 Sagitta en ata Sagitta minima Sagitta setosa Sagitta serratodentata September 199 March May June 199 July 199 September 199 March May July 199 September 199 Copepoda I II III IV Total Cladocera I II III IV Total Unidenti ed I food II III IV Total Total prey items March March Feeding and predation impact of chaetognaths G. Kehayias et al. 152

4 1528 G. Kehayias et al. Feeding and predation impact of chaetognaths Figure 2. Food containing ratio (FCR) values (means SE) of the developmental Stages (I^IV) of Sagitta en ata found in Strymonikos and Ierissos Gulfs during the ve sampling periods. sampling periods was calculated by multiplying the daily ingestion rate for each species by their mean density in the area. The numbers of copepods/cladocerans consumed m 3 d 1 by each chaetognath species were: P m 3 d 1 (the percentage of copepods/cladocerans in the diet of each chaetognath species)/100. The predation impact of chaetognaths on the copepod and cladoceran community, as percentage of the standing stock consumed daily, was estimated as: (number of copepods/cladocerans preyed per m 3 d 1 100)/number of copepods/cladocerans per m 3. Data on the mean abundance of copepod and cladoceran in the studied area were available for only three of the ve sampling periods ( July, September 199 and March ) and were obtained from Michaloudi (1999). RESULTS Abundance and stage composition of chaetognaths The highest value of the average abundance of the total chaetognaths was recorded in September, while the lowest was recorded in June (Table 1). Sagitta en ata (Grassi, 1881) was the dominant species in June, July and September 199 (59.0, 9.9 and 9.6% of the total chaetognaths, respectively), while Sagitta minima (Grassi, 1881) was the dominant species in March and May (82. and 69.2%, respectively). Sagitta setosa (Mu«ller, 184) and Sagitta serratodentata (Krohn, 1853) always comprised less than 8% of the total chaetognaths, being more abundant in September 199 and March respectively. The highest proportion of the immature (Stage I) specimens of S. en ata was recorded in September 199 and of S. minima in March, while the highest proportion of the mature (Stage IV) specimens of the former species in March and of the latter in May (Table 1). Feeding Of the 592 chaetognath specimens examined in the whole sampling area, 900 contained food in their guts (mean total FCR¼0.119). From all specimens examined, 3860 were Sagitta en ata with 540 of them containing food (mean FCR¼0.140), 3381 were Sagitta minima (282 of them containing food, mean FCR¼0.083), 25 were Sagitta serratodentata (3 of them containing food, mean FCR¼0.284) and 60 were Sagitta setosa (5 containing food, mean FCR¼0.083). The maximum mean FCR values of the total chaetognaths were recorded in September 199 and the minimum in May and this was also the fact for S. en ata (0.159 and 0.039, respectively) (Figure 1). The mean FCR values for S. minima ranged between 0.02 in September 199 and 0.5 in July. At the latter period only one prey item was found inthe guts of the two S. minima and of the three specimens of S. setosa collected. This is the reason for the highest FCR values for both species in this sampling period (0.50 and 0.33, respectively). Fed specimens of S. serratodentata were found only in March and the mean FCR for this species was (Figure 1). The diet of all chaetognath species is presented in Table 2. Copepods had the greatest proportion in the diet of total chaetognaths (436 food items, 48.2%), followed by cladocerans (264 food items, 29.2%), while unidenti ed food items were recovered from the guts of 199 chaetognaths (22.0%). Among the other zooplankton groups only one decapod larva, one bivalve larva, one anchovy egg and two chaetognaths were recovered from a total of ve S. en ata specimens in September 199. Copepods were the most important prey in the diet of S. en ata in July 199 and especially in March and May (38.4, 95.6 and 43.% of the total prey items, respectively), while

5 Feeding and predation impact of chaetognaths G. Kehayias et al Table 3. Predation pressure (prey consumed m 3 d 1 ) of the total chaetognaths and the individual species in all sampling periods in the north Aegean Sea. Values are expressed as average standard error. Species June 199 July 199 September 199 March May Sagitta en ata Sagitta minima Sagitta setosa ^ ^ Sagitta serratodentata ^ ^ ^ ^ Total chaetognaths consumption Table 4. The estimated predation impact of the total chaetognaths and the individual chaetognath species, expressed as percentage (%) of the total copepod and cladoceran standing stock consumed per day, during the three sampling periods in the north Aegean Sea. Values are expressed as average standard error. July 199 September 199 March Species Copepoda Cladocera Copepoda Cladocera Copepoda Cladocera X XSE Sagitta en ata Sagitta minima Sagitta setosa Sagitta serratodentata ^ ^ ^ ^ ^ Total chaetognaths consumption cladocerans comprised the greatest part in its diet in September 199, when they accounted for 53.9% of the total prey items (Table 2). Stage I, II and III specimens of S. en ata fed more intensively in September 199, while Stage IV in March (Figure 2). In September 199 Stages I, II, III and IV of S. en ata comprised 65.0, 28.3, 5.4 and 1.4% of the total population, respectively (Table 1), and they consumed nearly the same percentages (69.0, 23.1, 4.9 and 2.0%, respectively) of the total food items (Table 2). For Stage I specimens the greatest part of these food items (3.3%) were cladocerans and especially juvenile individuals of Penilia avirostris, Dana, In March, Stage I specimens of S. en ata comprised almost 30% of the total population, but they consumed 15.6% of the total prey items (copepods), while Stages II, III and IV comprised 28.0, 25.2 and 1.1% of the total population respectively (Table 1), and preyed on copepods by nearly the same percentages (33.3, 28.9 and 1.8%, respectively). It could be said then, that in March Stage I specimens of S. en ata preyed less intensively on copepods, than on cladocerans in September 199. Copepods were also the most important prey in the diet of S. minima (93.3%) being present in all sampling periods and especially in March, when the highest number of prey was found in the guts (Table 2). Cladocerans comprised 1.8% of the total prey items of this species, while unidenti ed food accounted for 5% (Table 2). Copepods accounted for 56.% in the diet of Stage III specimens, while 2.3% in Stage II and 8.9% in Stage IV specimens. Only one prey item (copepod) was found in the gut of one Stage I specimen of S. minima. Only ve prey items were recovered from S. setosa specimens; three of them were cladocerans and two were copepods. Fed specimens of S. serratodentata were found only in March. The unidenti ed food accounted for 63.0% in the diet of this species, while copepods accounted for the remaining 3.0%. Predation impact The predation pressure (number of prey consumed m 3 d 1 ) of the total chaetognaths and the individual species in each sampling period are shown in Table 3. The highest value of prey consumed per day by the total chaetognaths was recorded in September 199 and was due to the contribution of Sagitta en ata. This species always hada high pressure on total prey (% in June, 86% in July, 99% in September, 41% in May), being only replaced by Sagitta minima in March and May (66.% and 59%, respectively). The average abundance of copepods in the whole sampling area was highest in March (948.9 ind m 3 ), while of cladocerans in September 199 (249.9 ind m 3 ). The estimated predation impact of the total chaetognaths on the copepod/cladoceran community (as % of the total standing stock consumed per day) was greatest in September 199 reaching 2.6 and 1.29%, respectively due to S. en ata (Table 4). Stage I of this species accounted for 91.2% and 90.8%, respectively of the total copepod and cladoceran standing stock consumed per day by the total population of S. en ata in this period. In July 199, Sagitta en ata was also the only species contributing to a predation impact of 0.40% on the copepod and 0.12% on the

6 1530 G. Kehayias et al. Feeding and predation impact of chaetognaths cladoceran community. In March, chaetognaths and especially S. minima had the lowest predation impact on the copepod and the cladoceran community (0.02 and 0.01%, respectively) (Table 4). DISCUSSION Abundance and stage composition of chaetognaths The mean abundance of the total chaetognaths in the whole sampling area ranged between 0.29 and 1.39 ind m 3. However, abundance of the smaller chaetognath species (e.g. Sagitta minima) was probably underestimated due to net loss through the 500-mm net gauze used (Krsinic & Lucic, 1994). The seasonal variation in the abundance of the total chaetognaths was strongly a ected by the presence of Sagitta en ata, which was the most abundant species in the area during June, July and September 199. Sagitta en ata has been found to dominate the chaetognath community in several coastal areas of the Mediterranean being more abundant in late summer and early autumn (Kehayias, 1996; Batistic, 2003). Sagitta minima was the second more abundant species in the area especially in March and May. This species has been found in low percentages in the north Aegean Sea by Furnestin (190), but it was the dominant species on the north-west coasts of the Aegean Sea (Kehayias, 1996). In the south-east Adriatic Sea the maximum abundance of this species is recorded in August (Batistic, 2003), as well as on the coasts of Rhodes (Kehayias, 1996). However, in the Patraikos Gulf (north Ionian Sea) the maximum abundance is recorded in March, while on the north coasts of Crete the maximum abundance is recorded in November (Kehayias, 1996). The presence of Sagitta setosa and Sagitta serratodentata in low numbers in the north Aegean Sea has been also reported by Furnestin (190) and Kehayias (1996). The maximum abundance of S. en ata and the highest proportion of the immature (Stage I) specimens in September 199, suggest an intense breeding of this species in late summer to early autumn. Also, the presence of many mature individuals in March and the high proportion of Stage I specimens two months later, probably indicate a second breeding period of this species in spring. The maximum abundance of S. minima in March and the high proportions of mature (Stage IV) specimens during this period as well as immature specimens in May, suggest an intense breeding of this species in spring. However, the presence of immature specimens in all sampling months for S. en ata and almost all months for S. minima suggests that these species can reproduce throughout the year. Batistic (2003) reported both species to reproduce year-round in the Adriatic, while Kehayias (1996) reported that S. en ata might reproduce throughout the year with a major breeding period in late summer and that S. minima has its major breeding period in spring in a coastal area of western Greece. Feeding The proportion of the total chaetognaths containing food in their guts (FCR¼0.119) in the present study was greater compared with those recorded in other coastal (0.089; 0.058) and pelagic areas (0.048) of the eastern Mediterranean (Kehayias et al., 1996; Batistic et al., 2003 and Kehayias, 2003, respectively). However, the present ingestion rates are probably underestimated, taking into account that some chaetognath species (e.g. Sagitta en ata) usually feed at night (Pearre, 194; Kehayias et al., 1996; Giesecke & Gonzalez, 2004), while all stations in the present study were sampled during the day. The area of the present study is among the most productive areas of Greek waters (Michaloudi, 1999) and probably chaetognaths respond to that high abundance of food by increasing their feeding intensity. Among the chaetognath species found Sagitta serratodentata had the maximum mean FCR value, followed by S. en ata. Sagitta serratodentata has also been reported to have the maximum mean FCR values among the other chaetognath species in coastal and pelagic waters of the eastern Mediterranean (Kehayias et al., 1996; Kehayias, 2003), while there are no other references about the feeding intensity of this species elsewhere. The values of FCR for S. en ata in the Mediterranean and other marine areas of the world vary considerably between 0.02 and (Feigenbaum, 1991; Kehayias et al., 1996; Duro & Saiz, 2000; Batistic et al., 2003; Kehayias, 2003; Giesecke & Gonzalez, 2004). The mean FCR value of Sagitta minima in the present study (0.083) can be compared with the FCR values reported by Pearre (194) for the western Mediterranean (0.065^ 0.138), but it was higher than the values reported by Batistic et al. (2003) for the Adriatic (0.005^0.011) and the mean FCR value reported by Kehayias (2003) for the pelagic waters of the eastern Mediterranean (0.006). The mean FCR value of Sagitta setosa in the present study was and it was higher than the values reported by Batistic et al. (2003) for the Adriatic (0.03^0.06). It is not new to say that copepods are the most important food of chaetognaths in many marine areas worldwide, while other zooplanktonic groups are scarce in their diet (Feigenbaum, 1991). In the present study this was true especially for S. minima in March and May. Pearre (194) and Batistic et al. (2003) found copepods to be the main prey of this species (8.1 and 85.6%, respectively), followed by cladocerans and chaetognaths (less than 1.2%). The present study results show that apart from copepods cladocerans also comprised a considerable source of food for S. en ata, which preferred to feed either on copepods or cladocerans depending on their abundance. In the area of the present study, the abundance of cladocerans in July and September 199 was three to ve times greater compared to copepods (Michaloudi, 1999). Thus, it seems that when cladocerans exceed the abundance of copepods, S. en ata turns to them as the main source of food. However, in a seasonal study of the diet of chaetognaths in an oligotrophic coastal area of northern Crete, Kehayias et al. (1996) did not nd cladocerans to exceed the abundance of copepods and reported that the greatest proportion of cladocerans in the diet of S. en ata was 18% in September. Batistic et al. (2003) also found that the highest proportion of cladocerans in the diet of this species was not exceeding the 11.4% in September. The juvenile stages of cladocerans seem to have played a very important role in the diet of the new generation (Stage I specimens) of S. en ata. On the other hand, Stage I specimens showed less preference for copepods in March, than for cladocerans in September

7 Feeding and predation impact of chaetognaths G. Kehayias et al This could possibly be due to the di erence in size, in shape or in the swimming pattern of these two zooplanktonic groups, but it is not easy to explain since the copepods found in the gut contents were not classi ed to species level. Although S. en ata is considered as a good representative of the cannibalistic chaetognaths (Feigenbaum, 1991), this was not the case in our study considering the presence of only two chaetognaths in the gut content of S. en ata (0.4%). In previous studies, Kehayias et al. (1996) and Kehayias (2003) reported higher percentages of cannibalism for this species in the eastern Mediterranean (3.2 and 1.3%, respectively). Batistic et al. (2003) found pronounced cannibalism (.1^20.4%) for S. en ata, but also pronounced cannibalism for S. minima. Predation impact One of the di culties in the estimation of the predation impact of chaetognaths is the underestimation of the quantity of food by the use of a tow net (Feigenbaum, 1991). This could be the case for the juveniles of the cladoceran Penilia avirostris, which dominated the diet of Stage I specimens of S. en ata in September 199, although they were not so abundant in the samples (Michaloudi, 1999). In this case, there could be an overestimation of the predation impact of this species on the actual standing stock of cladocerans in this season. The predation impact of the total chaetognaths on the copepod and cladoceran standing stock in the studied area, ranged between 0.02^2.6% and 0.01^1.29% of the total copepods and cladocerans per day, respectively. These values are lower than those reported by Kehayias (2003) for the pelagic waters of the eastern Mediterranean, but higher compared with those reported by Duro & Saiz (2000) for the western Mediterranean. Sagitta en ata was the most important predator of both copepods and cladocerans in July and September 199, accounting for almost 100% of the predation impact of the total chaetognaths, while this was so for S. minima in March. Giesecke & Gonzalez (2004) o Chile found the daily predation impact of this species on the copepod standing stock to vary seasonally between 0.4% in winter and 6% in spring. However, they did not adjust their calculations by multiplying by two in the ingestion rate equation. In the only other reference about the predation impact of S. minima, Kehayias (2003) reported a maximum percentage of 0.3% of the copepods standing stock consumed per day in the pelagic waters of the eastern Mediterranean. Stage I of S. en ata accounted for more than 90% of the total copepod and cladoceran standing stock consumed per day by the total population of this species in September 199. This could be due to the maximum energy requirements of the juveniles of this species resulting from the increased summer reproduction. There are no relevant references for this species concerning the seasonal contribution of the di erent stages of development to the prey removal. The area of Strymonikos and Ierissos Gulfs is considered an important shing ground especially for the anchovy Engraulis encrasicolus (Linnaeus, 158) and sardine Sardina pilchardus (Walbaum, 192) of the north Aegean Sea (Kallianiotis, 1996). Also, Koutrakis et al. (2005) found anchovy larvae to dominate the total ichthyoplankton in June, July and especially September 199. Anchovy is considered as one of the main zooplanktivorous sh feeding mainly on copepods and other zooplankters, being able to have a daily food consumption of 1.4^23.8% of their body weight (Plounevez & Champalbert, 2000). Unfortunately, there are no data on the feeding of anchovy in the area of study to be compared with our results on feeding of chaetognaths. However, the latter could probably play an important role as competitors to the former, especially in certain periods (e.g. summer ^ early autumn) when the reproduction of anchovy occurs. In conclusion, the results of the present study are in agreement with the current opinion about chaetognaths being signi cant predators of the marine zooplankton, although there is no information for the contribution of other zooplankton predators in the area of our study. However, in this marine area of high productivity, chaetognaths exercised less pressure on their prey populations than in other oligotrophic areas of the eastern Mediterranean. The present study showed that copepods remain one of the main food sources for chaetognaths, but, for some species (e.g. Sagitta en ata) cladocerans could also play an important role to the energy transfer in speci c seasons, such as the reproductive periods. The study is part of a European Union co-funded LIFE project entitled Concerted actions for the management of the Strymonikos coastal zone, which was carried out by the Fisheries Research Institute of Kavala, Greece. REFERENCES Baier, C.T. & Purcell, J.E., 199. E ects of sampling and preservation on apparent feeding by chaetognaths. Marine Ecology Progress Series, 146,3^42. Bajkov, A.D., How to estimate the daily food consumption of sh under natural conditions. Transactions of the American Fisheries Society, 65, 288^289. Batistic, M., Abundance, biomass, C- and N- content of Flaccisagitta en ata and Mesosagitta minima (chaetognatha). Marine EcologyöPubblicazioni della Stazione Zoologica di Napoli, 24,1^13. Batistic, M., Mikus, J. & Njire, J., Chaetognaths in the South Adriatic: vertical distribution and feeding. Journal of the Marine Biological Association of the United Kingdom, 83, 1301^1306. Duro, A. & Saiz, E., Distribution and trophic ecology of chaetognaths in the western Mediterranean in relation to an inshore^o shore gradient. Journal of Plankton Research, 22, 339^361. Feigenbaum, D.L., Food and feeding behaviour. In The biology of chaetognaths (ed. Q. Bone et al.), pp. 45^54. Oxford: Oxford University Press. Furnestin, M.-L., 190. Cheatognathes des campagnes du Thor (1908^1911) en Me diterrane e etenmernoire. Dana Reports, 9,1^51. Giesecke, R. & Gonzalez, H.E., Feeding of Sagitta en ata and vertical distribution of chaetognaths in relation to low oxygen concentrations. Journal of Plankton Research, 26, 45^486. Kallianiotis, A., The anchovy shery in the Aegean Sea. A ourishing industry or a lost a air? Scientia Marina, 60, 28^288.

8 1532 G. Kehayias et al. Feeding and predation impact of chaetognaths Kehayias, G., Contribution on the study of the biology and ecology of chaetognaths in Greek waters. PhD thesis, University of Patras, Patra, Greece. [In Greek, English abstract.] Kehayias, G., Quantitative aspects of feeding of chaetognaths in the eastern Mediterranean pelagic waters. Journal of the Marine Biological Association of the United Kingdom, 83, 559^ 569. Kehayias, G., Koutsikopoulos, C., Fragopoulu, N. & Lykakis, J., A single maturity classi cation key for ve common Mediterranean chaetognath species. Journal of the Marine Biological Association of the United Kingdom, 9,113^1138. Kehayias, G., Lykakis, J. & Fragopoulu, N., The diets of the chaetognaths Sagitta en ata, S. serratodentata atlantica and S. bipunctata at di erent seasons in Eastern Mediterranean coastal waters. ICES Journal of Marine Science, 53, 83^846. Koutrakis, E.T., Kallianiotis, A.A. & Tsikliras, A.C., Temporal patterns of larval sh distribution and abundance in a coastal area of northern Greece. Scientia Marina, 68, 585^595. Krsinic,F.&Lucic, D., Mesozooplankton sampling experiments with the Adriatic Sampler: di erences of catch between 250 and 125 mm mesh netting gauze. Estuarine, Coastal and Shelf Science, 38, 113^118. Michaloudi, E., Mesozooplankton in the Strymonikos and Ierissos Gulfs. Description of the coastal zone of Strymonikos and Ierissos Gulfs (ed. E. Koutrakis and E. Lazaridou), pp. 441^453. NAGREFöFisheries Research Instituteö Greek Biotope/Wetland Centre, Nea Peramos, Kavala. [In Greek, English abstract.] Pearre, S. Jr, 194. Ecological studies of three west-mediterranean chaetognaths. Investigacion Pesqueras, 38, 325^369. Plounevez, S. & Champalbert, G., Diet, feeding behaviour and trophic activity of the anchovy (Engraulis encrasicolus L.) in the Gulf of Lions (Mediterranean Sea). Oceanologica Acta, 23, 15^191. Stamatis, N., Ioannidou, D. & Koutrakis, E., Monitoring of key eutrophication parameters at three inshore stations of Strymonikos Gulf, north Aegean Sea. Fresenius Environmental Bulletin, 10,06^10. Stuart, V. & Verheye, H.M., Diel vertical migration and feeding patterns of the chaetognath Sagitta friderici, o the coast of South Africa. Journal of Marine Research, 49, 493^515. Sylaios, G., Koutrakis, E. & Kallianiotis, A., Hydrographic variability, nutrient distribution and water mass dynamics in Strymonikos Gulf (N. Greece). Continental Shelf Research, inpress. UNESCO, Zooplankton sampling. In Monographs of oceanographic methodology, vol. 2 (ed. J.H. Fraser), pp. 1^14. Paris: UNESCO Press. Submitted 14 January Accepted 2 September 2005.