• Distribution;
  • habitat;
  • Hippocampus ;
  • underwater visual census


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Population abundance, distribution and habitat preference of the Mediterranean sympatric seahorses Hippocampus guttulatus and Hippocampus hippocampus were investigated in a semi-enclosed sea system (Apulian coast, Ionian Sea). A total of 242 individuals of seahorses were sighted in the 11 transects surveyed in summer 2011. Hippocampus guttulatus (n = 225) were 14 times more abundant than H. hippocampus (17). The mean abundance of H. guttulatus for all the pooled sites was 0.018 m−2 (SE ± 0.003) ranging from a maximum of 0.035 (SE ± 0.007) to a minimum of 0.008 (SE ± 0.002). The size structure of long-snouted seahorse shows a population ranging from 7 to 14 cm (SL) with a peak at 10 cm (TL). Juveniles (96.0 ± 8.0 mm) represent a significant fraction of the population, accounting more than 21% of the sighted individuals. In Mar Piccolo, H. guttulatus is able to shelter both in monotonous habitats, including the algal beds, and diversified ones, such as the rich filter-feeder communities that colonize hard substrates. By contrast, H. hippocampus is mainly associated with habitats of low complexity. Today, the Mar Piccolo di Taranto is among the most heavily polluted water bodies in South Italy, with trace metals, hydrocarbons, pesticides and organic wastes affecting both biotic and abiotic matrices. However, despite the high level of degradation, the presence of a large mussel farm has avoided the impact of towed fishing gears, and eutrophication of water bodies has ensured a high trophic level that supports large crustacean populations, potential prey for seahorses.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Worldwide, seahorse (Hippocampus Rafinesque 1810 spp.) populations and their relatives in the family Syngnathidae (pipefishes, seadragons and pipehorses) are threatened by (i) degradation of their estuarine, seagrass, mangrove and coral habitats (Olden et al. 2007), (ii) incidental capture in fishing gear (bycatch) (Vincent 1996), and (iii) over-exploitation for use in traditional medicines and in the aquarium trade (Salin et al. 2005). Moreover, seahorses have life histories, behaviour and ecology that make them particularly vulnerable to population decline (Foster & Vincent 2004): they provide lengthy and vital parental care for small broods, exhibit low mobility and high site fidelity, have low natural rates of adult mortality, and (in many species) maintain faithful pairbonds (Lourie et al. 1999; Vincent et al. 2011). In addition, seahorses inhabit shallow, coastal areas worldwide, where anthropogenic disturbances tend to be the most frequent and severe, and represent the main threat to their integrity. The decline of seahorse populations worldwide has therefore been brought to the attention of the international community, leading to their classification as threatened species, and their inclusion on the World Conservation Union Red List of Threatened Species (Vincent & Hall 1996; World Conservation Union 2002) and, in 2002, the Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES 2002).

In particular, Hippocampus guttulatus Cuvier, 1829 and Hippocampus hippocampus Linnaeus, 1758 – the two Mediterranean species of seahorses – are also listed under OSPAR, European CITES (Curd 2009), the Bern Convention and the Barcelona Convention (Abdul Malak et al. 2011) and protected by the UK Wildlife and Countryside Act of 2008 (DEFRA 2008). However, as few data are available across most of their geographic range, neither species can be reliably classified in terms of global level risk according to IUCN categories, and they are listed, at present, as data-deficient (DD).

Along the Italian coasts, the trend of Hippocampus populations (H. guttulatus and H. hippocampus) is very similar to that described for other seahorse species in coastal areas worldwide (Vincent et al. 2011). Until 80, sightings of seahorses along Italian coasts were frequent, with individuals of both species often being accidentally taken as bycatch by trammel nets and trawlers. However, in the last few decades the degradation of sensitive habitats, the use of illegal fishing gear on coastal biocenoses, and the development of snorkelling and SCUBA diving have slowly but inexorably reduced seahorse population densities, structure and spatial distribution. As a result, in the last decade, seahorse sightings have been rare or scanty along most of the Italian coast and populations appear scattered, showing peaks of abundance in the Northern Tyrrhenian (Tuscany and Liguria) and Northern Adriatic Seas (Goffredo et al. 2004). Paradoxically, species which have been included on the World Conservation Union Red List of Threatened Species and are protected by several international conventions (CITES 2002) appear to have found suitable habitats in heavily anthropic areas. Indeed, Hippocampus spp. populations now appear to be mainly concentrated in harbours, mussel farms or artificial structures, which provide shelter from the direct impact of mechanical human activities (commercial fishing, dredging, dumping, SCUBA diving, snorkelling).

In this paper, we describe the Hippocampus populations (H. guttulatus and H. hippocampus) in the semi-enclosed sea system around the city of Taranto (Apulian coast, Ionian Sea). The first scientific report of seahorses in the Mar Piccolo of Taranto dates back to Cerruti (1938). Later, Parenzan (1969) reported the occurrence of Hippocampus in association with photophilous algal beds (mainly ‘Caulerpa, Rhytiphloea, Gracilaria spp., Chaetomorpha and Cladophora prolifera’). According to older local fishermen, in the last century, seahorses were frequently sighted and collected as bycatch during traditional fishing activities. Moreover, fishermen report a remarkable increase in seahorse sightings in the last 50 years, with the spread of mussel farming. According to this scenario, the increase in pools and hard structures utilized by farmers had strongly reduced the fishery, thus reducing sea horse captures. Purse seining, which up to the middle of the 20th century was the prevalent fishing activity within the Mar Piccolo, is now only very occasionally used, and is confined to a few tracts along the eastern coast of the inner inlet (Fig. 1).


Figure 1. Mar Piccolo di Taranto. Location of 11 underwater visual census sites: 1– 8 subtidal sites 0.5 – 5 m depth; A-C deep sites (5.1 – 10 m depth). Surface area per stratum is also indicated.

Download figure to PowerPoint

Today, the Mar Piccolo di Taranto is among the most heavily polluted water bodies in Southern Italy. Surrounding urban developments are known and presumed to have affected the semi-enclosed sea ecosystem with trace metals, hydrocarbons, pesticides and organic wastes (Cardellicchio et al. 2007; Petronio et al. 2012). However, despite the high level of degradation, the presence of a large mussel farm avoids any impact from towed fishing gears and the eutrophication of the water bodies ensures a high trophic level that supports large crustacean populations, mainly amphipods, which are potential prey for seahorses (Prato & Biandolino 2003).

In the present paper, our main objectives were to (i) assess the abundance of the local population; (ii) evaluate differences in habitat preference; (iii) describe the spatial distribution of Hippocampus populations in the semi-enclosed sea system of the Mar Piccolo. To achieve these goals, a visual census survey (during summer 2011) was carried out at 11 stations representative of the main benthic habitats and distributed along a confinement gradient. Abundance of the two species was then compared across different stations and substrates, and habitat preference was established.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

Study species

The long-snouted (Hippocampus guttulatus) and the short-snouted (H. hippocampus) seahorse are notable residents of European coastal waters. Both species inhabit shallow inshore waters, mainly in seagrass meadows and algal beds from the Eastern Atlantic to the Western Mediterranean (Lourie et al. 2004).

Hippocampus hippocampus occurs in shallow coastal waters from the North Sea, down the European coast and into the Mediterranean Sea. Its geographic range extends down the Atlantic African coast as far as Senegal, including the Atlantic islands of the Azores, Madeira and Canaries (Lourie et al. 2004). Exact locations for the populations are unknown, but it is expected that seahorses usually occur at low densities along the coastline (L. Woodall personal observations).

Hippocampus guttulatus is present along the Atlantic coast from the UK, Ireland and Netherlands to the Mediterranean Sea. Distribution continues into the Mediterranean Sea and the Black Sea (Lourie et al. 2004). Although the two species are commonly distinguished by the presence or absence of skin filament, having a smooth or a filamentous appearance has proved to be an unreliable distinction. Otherwise, H. guttulatus and H. hippocampus can be identified (also in situ) by differences in head, snout and trunk shape (Lourie et al. 1999).

Study area description

The Mar Piccolo of Taranto (40°28′N, 17°16′W) is located at the northern end of the Gulf of Taranto (Ionian Sea). It has a total surface area of about 20.63 km2 and consists of a system of two distinct inlets, called ‘Primo’ and ‘Secondo Seno’ (first and second inlet), the innermost of which receives an inflow from some small streams (Fig. 1). The Mar Piccolo is characterized by low water turnover and water movement. Water circulation is mainly due to water flowing through the main channel, supplied by tidal currents and by a powerful pump utilized for industrial purposes. Currents are moderate, anticyclonic, and more intense in the first inlet than in the second one, where they are more appreciable near the narrows which connect the two inlets (Capasso et al. 1989). Water temperature varies from 8.4 °C (December) to 26.8 °C (August). Salinity values are around 36‰ (Caroppo & Cardellicchio 1995). Locally, however, continental inflows may determine a marked drop in salinity (25–20‰) (Corriero G. & Cardone F., unpublished data). Dissolved oxygen shows wide seasonal variability: from May to October, the oxygen deficit in the deep layers of the Mar Piccolo (from about 7 to 10 m in depth) can reach marked levels which are responsible for dystrophic events (Alabiso et al. 1997; Gaino et al. 2010). Soft bottoms are dominant, varying from muds to mixed sands. They are sparsely covered by patches of Cymodocea nodosa, currently reported to be in regression, and by large beds of unattached macroalgae (Cecere & Petrocelli 2009). Where plants are missing, large numbers of sabellids, molluscs and ceriantharia have colonized the soft bottoms. Hard substrates are very abundant and diversified, consisting mainly of stones and debris of human origin, ropes and material abandoned by mussel farmers and fishermen. In addition, there are several kilometres of stone walls along the coastline. The macrozoobenthic community is rather unique, characterized by very rich assemblages of filter feeders (mainly poriferans, hydrozoans, polychaetes, bryozoans, bivalves, crinoids and ascidians) that colonize all of the various types of hard substrate, forming pluristratified assemblages. Among the vagile fauna, amphipods – in particular Gammarus aequicauda – are dominant, with remarkable seasonal peaks. According to the literature (Kevrekidis & Koukouras 1989), this species – which in summer in the Mar Piccolo reaches densities of up to 7200 individuals per m2 (Prato & Biandolino 2003) – plays a key role in the structure of benthic communities and is an important component in marine benthic food webs. The zooplanktonic community is rich, diversified and subject to seasonal changes in biomass, with autumn peaks (Belmonte et al. 2001).

Zooplankton composition and distribution shows a confinement gradient from the outer to the inner area (Belmonte et al. 2001) according to the model of paralic systems (Guelorget & Perthuisot 1992). A total of 119 taxa belonging to 15 phyla were surveyed in the Mar Piccolo; the abundance of zooplankton ranged from 53,917 ind.·m−3 in the first inlet to 2720 ind.·m−3 in the second inlet (Belmonte et al. 2001).

Underwater visual census

A preliminary survey was carried out in summer 2011 (June–September) to obtain a general overview of the Mar Piccolo's habitat features and spatial distribution of the Hippocampus spp. As a consequence, in summer 2012 (from June to September) a total of eight sites were surveyed by SCUBA diving (Fig. 1). Surveys were carried out using standard underwater visual census (UVC) techniques (Samoilys 1997). For each site, two stations were set up, and for each station two randomly selected 2 × 150 m linear transects were monitored. The transects were located perpendicular to the coastal line at a depth of between 0.5 and 5 m, and were >50 m apart. Although during the preliminary survey the bulk of seahorse individuals was concentrated in the subtidal area (0.5–5 m depth), to experimentally verify the bathymetric range occupied by Hippocampus spp. populations, three additional sites (A–C) were chosen at depths of between 5.0 and 10.0 m. This sampling design enabled a total area of about 26,400 m2 to be surveyed.

For each seahorse sighted on the transects, sex, total length and life stage (juvenile or adult) were recorded in situ; to obtain further biometric information, all specimens encountered were photographed with a ruler placed as close to them as possible (Curtis et al. 2004). Hippocampus guttulatus specimens were considered juveniles if they were <50% size at maturity, corresponding to 96.0 ± 8.0 mm (TL) (Curtis 2004). Moreover, for each sighting, information on depth and habitat was recorded. Data on population density used in the subsequent analysis were averaged among transects within sites. Finally, the abundance of homogeneous habitat structures was estimated as percent cover of the surveyed area for each transect. Using sedimentological and biological features of the sea bottom, SCUBA divers described and mapped (along each transect) the following different habitats:

  • Pools with soft bottom (PWSB) – Wood pools are colonized by a rich sessile filter feeder fauna, with more than 100 species of zoobenthic invertebrates reported in the literature (Scalera-Liaci & Corriero 1993; Longo et al. 2004; Panetta et al. 2004; Mastrototaro et al. 2008; Pierri et al. 2010). This benthic assemblage is distributed in a pluristratified pattern. The basal layer is mainly colonized by bivalves (Mytilus galloprovincialis and Ostrea edulis), demosponges (Cliona spp., Hymeniacidon perlevis, Oceanapia isodyctiformis, Suberites carnosus) and ascidians (Styela spp.). The epibenthic layer is characterized by various benthic facies, varying with depth. From the surface to about 1 m depth, macroalgae and sea anemones (Anemonia sulcata) largely prevail. From 1 m to about 3 m depth, the sabellids Sabella spallanzani and Branchiomma luctuosum, together with a dense community of soft hydrozoans and a few large encrusting colonial didemnidae, are dominant. At greater depths, the epibenthic layer species mainly consist of large specimens of the calcareous sponge Paraleucilla magna, serpulids (Hydroides spp.), large bioconstructions of the bryozoan Schizoporella sanguinea and large ascidians (Microcosmus spp., Phallusia mammillata). Due to the usual occurrence of large amounts of hard artificial substrates (ropes, chains, bricks) and benthic organisms (mussel shells, bryozoa), the soft bottom displaced around each pool was also sampled.
  • Artificial hard substrates (AHS) – These substrates are colonized by macroalgal assemblages down to a depth of about 1 m; at the greatest depth they are heavily colonized by bivalves, gastropods (Murex sp.), large sabellids, both colonial and solitary ascidians, demosponges, bryozoans and hydrozoans.
  • Vegetal beds Cymodocea nodosa (CYNOD), Ulva spp. (ULVA), Chaetomorpha linum (CLIN), Cladophora prolifera (CPRO). Cymodocea nodosa is the only phanerogam reported for the Mar Piccolo (Pierpaoli 1923; Cecere & Petrocelli 2009), with several scattered patches, varying greatly in size and density, occurring in both the inlets (Cecere & Petrocelli 2009). Seaweed beds consist of very monotonous habitats, with a few associated zoobenthic species (some gastropods, large opistobranchs, the sea anemone Aiptasia sp.), mainly distributed in the upper layer of the phytocoenosis. A very rich vagile fauna, with a few species of amphipods and isopods, is usually associated with such algal beds.
  • Soft bottom with scattered concretions (SBSB) – Near the coastline, soft bottoms are mixed with a large amount of organogenous concretions (bivalve and gastropods shells), small stones and artificial hard substrates, which are mainly colonized by sabellids and solitary ascidians.

Data analysis

To assess patterns of seahorse habitat preference in the Mar Piccolo subtidal area, a two-way design was used, including the following factors: ‘Habitat’ as a fixed factor with seven levels – i.e. soft bottoms with scattered bioconcretions (SBSB), Cymodocea nodosa beds (CYNOD), Chaetomorpha linum beds (CLIN), Ulva beds (ULVA), pools with soft bottom (PWSB), Artificial hard substrates (AHS) and Muds (MUDS) – and ‘Sites’ as a fixed orthogonal factor with eight levels (Sites 1–8). Four random density replicates (number of individuals · 1 m−2) were considered for each combination of the above factors (n = 224). When ANOVA gave significant results, a post hoc comparison was performed using the Student–Newman–Keul (SNK) test. All the analyses were performed using GMAVinc. Software (Underwood 1997).

On the basis of a bathymetric map (1: 20,000) provided by the Italian Military Geographic Institute and available from the Cartographic Service of the Apulia Region (, the surface per stratum of the Mar Piccolo was calculated in both the inlets. In particular, using an ARCVIEW GIS ESRI 3.2 platform, the surfaces of the subtidal (0–5 m), intermediate (5.1–10 m) and deeper layers (>10 m) were estimated. Population sizes were estimated by extrapolating the mean densities of the subtidal and intermediate layer surfaces.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

A total of 242 individuals of seahorses were sighted along the 11 transects surveyed. Hippocampus guttulatus (n = 225) was 14 times more abundant than H. hippocampus (n = 17) (Table 1). Hippocampus guttulatus specimens were distributed at all sites surveyed, whereas H. hippocampus individuals were located only at sites 3, 4 and 6 (see Fig. 1). The mean abundance of H. guttulatus for all pooled sites was 0.018 per m2 (SE ± 0.003), ranging from a maximum of 0.035 (SE ± 0.007) at site 3 to a minimum of 0.008 (SE ± 0.002) at site 1.

Table 1. Indices of abundance for Hippocampus guttulatus and Hippocampus hippocampus
 Mar Piccolo subtidal sitesMar Piccolo deep sitespooled sites
Hippocampus guttulatus
Total no.83128253422910441176
Mean density (m−2)0.0080.0330.0350.0310.0330.0230.0080.0100.0060.0110.0060.018
Hippocampus hippocampus
Total no.00130120000016
Mean density (m−2)0.0000.0000.0010.0030.0000.0010.0000.0000.0000.0000.0000.0005

In particular, H. guttulatus density was higher at sites in the central area of the Mar Piccolo (sites 2, 3, 4, 5) (Fig. 1, Table 1), and less abundant in the outer and inner parts of the semi-enclosed sea. The H. hippocampus mean density for all the pooled sites was, obviously, very low (0.0005 ± 0.0002 SE) with a peak at site 6 (0.003 ± 0.001 SE) located in the inner part of the Mar Piccolo (Fig. 1). The H. guttulatus sex ratio (females/total) was slightly female-biased (53.7%); sightings of H. hippocampus were too limited to calculate a proportion by sex.

The size (sexes combined) of long-snouted seahorse shows a population ranging from 7 to 14 cm (TL) with a peak at 10 cm (TL). Juveniles (with a size <96.0 ± 8.0 mm) represent a significant fraction of the population, accounting for more than 21% of the sighted individuals (Fig. 2). Unfortunately, there were too few sightings of H. hippocampus for a proper analysis of the population structure.


Figure 2. Standard length-frequency distribution (sex combined) of individuals of H. guttulatus sighted from June to August 2012 (10 mm size classes).

Download figure to PowerPoint

Hippocampus guttulatus density (data pooled across all sites) was broadly related to habitat characteristics. Ulva beds shelter the maximum density of long-snouted seahorse (0.058 ind.·m−2), whereas no sightings were recorded on Muds. The two-way ANOVA analysis performed on H. guttulatus density revealed significant differences between ‘Sites’ and ‘Habitats’ (Table 2). The SNK test performed on the ‘Habitat × Sites’ combination shows no differences in habitat preference in sites 1 and 7 (Table 2, Fig. 1). ULVA was the favourite habitat at sites 2 and 5, whereas artificial structures (AHS and PWSB) were the preferred habitats at sites 3, 4 and 8 (Table 2, Fig. 3).

Table 2. Summary of the results of the Student–Newman–Keuls post hoc tests. Only significant interaction terms (P < 0.01) are reported
  1. ***P < 0.0001.

Sites70.22760.032525 14***
Habitat × Sites420.90920.021616.74***
Habitat × sites SNK test
1No differences
3AHS = PWSB > all
4AHS > PWSB = SBSB > all
5ULVA > PWSB > all
6PWSB > all
7No differences
8PWSB > all

Figure 3. Density of Hippocampus guttulatus in the surveyed habitats.

Download figure to PowerPoint

As reported in Fig. 1, the first inlet covers a total area of 8.01 km2, with 1.19 km2 within the bathymetric range 0–5 m, 1.93 km2 within the bathymetric range 5.1–10 m and 4.89 km2 deeper than 10 m. The second inlet covers a total area of 12.60 km2, with 3.98 km2 within the bathymetric range 0–5 m, 8.39 km2 within the bathymetric range 5.1–10 m and 0.25 km2 deeper than 10 m. Thus, considering the whole Mar Piccolo, the subtidal layer (0–5 m) covers a total of 5.17 km2, whereas the 5–10 m layer measures about 10.32 km2.

The estimate of population size shows a very high number of H. guttulatus individuals in the subtidal area [116,325 individuals, lower 95% confidence interval (CI) = 76,029; upper 95%CI = 156,621] and a less abundant fraction of the population in the intermediate stratum (5–10 m) (80,267 individuals, lower 95%CI = 47,316; upper 95%CI = 113,217). The H. hippocampus population is significantly small in the subtidal layer (3231 individuals, lower 95% CI = 1,878; upper 95%CI = 6,837) and completely absent in the intermediate layer.


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References

It is well documented that several seahorse species worldwide are targeted by fisheries in several countries (e.g. Vietnam, India), where they are dried for traditional medicine markets (Zhang et al. 2003), for curiosities (Grey et al. 2005) or sold alive for the aquarium market (Martin-Smith & Vincent 2006). Moreover, several aspects of the seahorse's life strategies (i.e. small body size, paternal care, low mobility, sedentary adults and monogamous mating) (Foster & Vincent 2004) leave them particularly vulnerable to disturbance from human activities and sensitive to habitat degradation. European seahorse populations (Hippocampus hippocampus and Hippocampus guttulatus) in particular, which are not specifically targeted by fishing activities, are mainly threatened by accidental catch in towed fishing gears and by their sensitivity to habitat degradation (Woodall et al. 2011). According to the literature, H. guttulatus and H. hippocampus are considered to be organisms found in coastal habitats or ecosystems with good environmental status, very sensitive to environmental conditions and therefore a useful tool for analysing the environment health as a flagship species (Garrick-Maidment & Jones 2004). However, the present study describes seahorse populations (H. guttulatus and H. hippocampus) in a heavily polluted and human impacted semi-enclosed sea in Southern Italy. The estimated total abundance of seahorses in the two inlets of the Mar Piccolo di Taranto is really impressive, with values of 123,456–269,838 and 1878–6837 for H. guttulatus and H. hippocampus, respectively. Moreover, given the cryptic behaviour of seahorses and the poor water clarity in the study area, the mean density and population size of seahorses could be greatly underestimated. The long-snouted seahorse density recorded in the present study was comparable to that of similar sites in the Atlantic Sea (Curtis & Vincent 2005; Caldwell & Vincent 2012) and greater than that reported in the Mediterranean Sea (Goffredo et al. 2004), the latter being obtained by counting the number of seahorses per dive-hour. Length–frequency distribution (sex combined) of H. guttulatus shows a young population as whole, with a lack of large adults (more than 16 cm TL), and juveniles accounting for more than 21% of the population. Although a recruitment pattern is widely expected in the summer period, the juvenile fraction in the Mar Piccolo is abundant and, according to the literature (Woodall 2009), this would seem to reflect a good population status.

In the last three decades, the original landscape and ecology of the Mar Piccolo have been radically modified; soft bottoms have become dominant, varying from muds to mixed sands, whereas once-dominant sea grasses (Cymodocea nodosa) are now reported to be in regression, scattered in sparse patches and largely replaced by beds of unattached macroalgae such as Ulva spp. and Chaetomorpha linum (Cecere & Petrocelli 2009). Moreover, since the 1970s, mussel farms have colonized most of the sea bottom surface, increasing the availability of hard substrates and forming a mechanical obstacle to fishing activities with towed gears. Despite the critical environmental conditions, the Mar Piccolo di Taranto shelters a wide population of seahorses. In agreement with the data of Ria Formosa lagoon (Curtis & Vincent 2005), the population of H. guttulatus appears to be more numerous and distributed over a wider area, whereas sightings of H. hippocampus are scanty and concentrated in particular areas. According to Curtis & Vincent (2005), H. hippocampus in Mar Piccolo is mainly associated with habitats of low complexity (site 6; Fig. 1), whereas in other sites with the greatest habitat heterogeneity, its presence is rare and scanty. By contrast, H. guttulatus is able to shelter both in monotonous habitats, such as algal beds, and in diversified ones, such as the rich filter-feeder communities that colonize hard substrates. In addition, this seahorse can be found grasping both natural and artificial substrates, such as wooden poles, nets and ropes. The role of artificial substrates in sheltering seahorses has already been highlighted for both Mediterranean species (Curtis & Vincent 2005) and exotic species (Hellyer et al. 2011). Generally, seahorse density is patchy (Foster & Vincent 2004) and at most locations they are rare; however, they can be locally abundant (Curtis & Vincent 2005; Woodall 2009). Our data showed that seahorses are mainly concentrated in the central part of the Mar Piccolo (Sites 2, 3, 4, 5) (Fig. 1) where about 56% of the individuals were detected. The central part of the Mar Piccolo is served by tidal currents from the open sea (Capasso et al. 1989) and is unaffected by illegal activities of handling towed fishing gears practised in the eastern side of the second inlet. No confinement gradient has been observed, as the number of sighted specimens decreases from the central portion of the Mar Piccolo towards the innermost and the outermost portions of the sea system. The depth range seems to be more contracted than that reported in the literature for European seahorses. In fact, most of the sightings occurred in the subtidal zone, with only a small fraction of the population sheltered in the intermediate sites (5–10 m), where muds are dominant. This finding can be partially explained by the frequent dystrophic and anoxic events occurring in the deep layers of the Mar Piccolo (from about 7–10 m of depth) in summer time (Alabiso et al. 1997; Gaino et al. 2010). Moreover, the anoxia phenomenon seems to prevent the seasonal migration into deeper water (>10 m) described by Boisseau (1967) in cooler waters for both H. hippocampus and H. guttulatus (Corriero pers. obs.). In addition, most of the suitable seashore habitats, such as phanerogams and algal beds, but also hard substrates, are mainly concentrated within the first 5 m of depth.

Seagrasses (Posidonia oceanica, C. nodosa) are widely thought to be a common habitat for European seahorse species (Foster & Vincent 2004; Garrick-Maidment 2007). However, in the current study, only a very small number of seahorses were surveyed on the residual C. nodosa beds. Notwithstanding, in the Mar Piccolo di Taranto the seagrass has declined dramatically, and seahorse populations seem to prefer to shelter and feed in artificial habitats where they can find resources and safe holdfasts.

Hard complex substrates of human origin (mussel farm pools, rocky artificial substrates) are the most favourable habitats for seahorse populations in the Mar Piccolo. The abundant colonization of epifauna on artificial hard substrates offers shelter and camouflage and supports very abundant potential prey populations while increasing protection against predators (Wennhage & Pihl 2007). Both mussel farm pools and rocky artificial substrates increased structural complexity, maintaining a very high abundance of mobile crustaceans (mainly amphipods and copepods), which are considered to be common prey items for H. guttulatus and H. hippocampus (Kitsos et al. 2008). Moreover, this finding agrees with a recent study carried out by Hellyer et al. (2011) in Sydney Harbour, where artificial habitats support much higher abundances of the mobile epifauna (amphipods, copepods) that are consumed by seahorses. Among the algae, Ulva beds are a habitat of low complexity, giving rise to a rich detritus food chain due to its rapid turnover, thus supporting an abundant population of crustacean detritivores. By contrast, the presence of seahorse individuals on C. linum is rare or scanty; as suggested by Sundin et al. (2011), filamentous algae reduce visibility by obscuring objects in the water, thus decreasing the possibility of food intake.

Further study is required to determine whether habitat distribution changes on a seasonal basis or depending on the sex and life stages of seahorse individuals. In our opinion, in fact, whether prey item availability could be the main factor explaining the patchy distribution of seahorse habitat, ecological and behavioural aspects can affect the use and habitat preference during crucial and vulnerable stages of the seahorse life cycle (e.g. recruitment, courtship, coupling, male parental care).


  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. References
  • Abdul Malak D., Livingstone S.R., Pollard D., Polidoro B.A., Cuttelod A., Bariche M., Bilecenoglu M., Carpenter K.E., Collette B.B., Francour P., Goren M., Hichem Kara M., Massutí E., Papaconstantinou C., Tunesi L. (2011) Overview of the Conservation Status of the Marine Fishes of the Mediterranean Sea. IUCN, Gland, Switzerland and Malaga, Spain: vii + 61 pp. ISBN: 978-2-8317-1307-6.
  • Alabiso G., Cannalire M., Ghionda D., Milillo M., Leone G., Caciorgna O. (1997) Particulate matter and chemical-physical conditions of an inner sea: the Mar Piccolo in Taranto. A new statistical approach. Marine Chemistry, 58, 373338.
  • Belmonte G., Fanelli G., Gravili C., Rubino F. (2001) Composition, distribution and seasonality of zooplankton in Taranto seas (Ionian Sea, Italy). Biologia Marina Mediterranea, 8, 352362.
  • Boisseau J. (1967) Les Regulations Hormonales de l'incubation chez un vertebre male: recherches sur la reproduction de l'hippocampe. PhD dissertation, Université de Bordeaux, Bordeaux, France.
  • Caldwell J.R., Vincent A.C.J. (2012) Revisiting two sympatric European seahorse species: apparent decline in the absence of exploitation. Aquatic Conservation Marine Freshwater Ecosystem, 22, 427435.
  • Capasso V., Di Liddo A., Notarnicola F., Posa D., Rinaldi F. (1989) Modelli matematici per il controllo della qualità delle acque costiere. Atti Convegno Nazionale Progetto Strategico CNR ‘Matematica computazionale’, Rome: 24.
  • Cardellicchio N., Buccolieri A., Giandomenico S., Lopez L., Pizzulli L., Spada L. (2007) Organic pollutants (PAHs, PCBs) in sediments from the Mar Piccolo in Taranto (Ionian Sea, Southern Italy). Marine Pollution Bulletin, 55, 451458.
  • Caroppo C., Cardellicchio N. (1995) Phytoplankton of Mar Piccolo of Taranto (Jonian sea). Oebalia, 21, 6176.
  • Cecere E., Petrocelli A. (2009) The Mar Piccolo of Taranto. In: Cecere E., Petrocelli A., Sfriso A., Izzo G. (Eds), Flora and Vegetation of the Italian Transitional Water Systems. CoRiLa, Multigraf, Spinea: 195227.
  • Cerruti A. (1938) Le condizioni oceanografiche e biologiche del Mar Piccolo di Taranto durante l'agosto del 1938. Bollettino Piscicoltura Idrobiologia, Roma: 711751.
  • CITES. (2002) Conservation of seahorses and other members of the family Syngnathidae. Twelfth Meeting of the Conference of the Parties, Santiago, Chile. Available at:
  • Curd A. (2009) Background document for the short-snouted seahorse Hippocampus hippocampus. OSPAR Commission, Biodiversity Series.
  • Curtis J.M.R. (2004) Life History, Ecology and Conservation of European Seahorses. McGill University Montréal Québec, PhD dissertation. Department of Biology: 1215.
  • Curtis J.M.R., Vincent A.C.J. (2005) Distribution of sympatric seahorse species along a gradient of habitat complexity in a seagrass-dominated community. Marine Ecology Progress Series, 291, 8191.
  • Curtis J., Moreau M.A., Marsden D., Bell E., Martin-Smith K., Samoilys M., Vincent A. (2004) Underwater visual census for seahorse population assessments. Project Seahorse Technical Report No.8, Version 1.0. Project Seahorse, Fisheries Centre, University of British Columbia. 28 pp.
  • DEFRA. (2008) UK Wildlife and Countryside Act of 2008. Available at:
  • Foster S., Vincent A.C.J. (2004) Life history and ecology of seahorses: implications for conservation and management. Journal of Fish Biology, 65, 161.
  • Gaino E., Cardone F., Corriero G. (2010) Reproduction of the intertidal sponge Hymeniacidon perlevis (Montagu) along a bathymetric gradient. Open Marine Biology Journal, 4, 4756.
  • Garrick-Maidment N. (2007) British Seahorse Survey Report. The Seahorse Trust, Topsham: 126.
  • Garrick-Maidment N., Jones L. (2004) British Seahorse Survey Report. The Seahorse Trust, Topsham: 183.
  • Goffredo S., Piccinetti C., Zaccanti F. (2004) Volunteers in marine conservation monitoring: a study of the distribution of seahorses carried out in collaboration with recreational scuba divers. Conservation Biology, 18, 14921503.
  • Grey M., Blais A.M., Vincent A.C.J. (2005) Magnitude and trends of marine fish curio imports to the USA. Oryx, 39, 413420.
  • Guelorget O., Perthuisot J.P. (1992) Paralic ecosystems. Biological organization and functioning. Vie Milieu, 42, 215251.
  • Hellyer C.B., Harasti D., Poore A.G.B. (2011) Manipulating artificial habitats to benefit seahorses in Sydney Harbour, Australia. Aquatic Conservation Marine Freshwater Ecosystem, 21, 582589.
  • Kevrekidis T., Koukouras A. (1989) Seasonal variation of abundance of Gammarus aequicauda (Crustacea: Amphipoda) in the Evros Delta (NE Greece). Israel Journal of Zoology, 36, 113123.
  • Kitsos M.S., Tzomos T., Anagnostopoulou L., Koukouras A. (2008) Diet composition of the seahorses, Hippocampus guttulatus Cuvier, 1829 and Hippocampus hippocampus (L. 1758) (Teleostei, Syngnathidae) in the Aegean Sea. Journal of Fish Biology, 72, 12591267.
  • Longo C., Scalera-Liaci L., Corriero G. (2004) Note sui poriferi del Mar Grande e del Mar Piccolo di Taranto (Mar Ionio). Biologia Marina Mediterranea, 11, 440443.
  • Lourie S.A., Vincent A.C.J., Hall H.J. (1999) Seahorses: an Identification Guide to the World's Species and their Conservation. Project Seahorse, London: 214.
  • Lourie S.A., Foster S.J., Cooper E.W.T., Vincent A.C.J. (2004) A Guide to the Identification of Seahorses. University of British Columbia and World Wildlife Fund, Washington, DC: 1114.
  • Martin-Smith K.M., Vincent A.C.J. (2006) Exploitation and trade of Australian seahorses, pipehorses, sea dragons and pipefishes (Family Syngnathidae). Oryx, 40, 141151.
  • Mastrototaro F., D'Onghia G., Tursi A. (2008) Spatial and seasonal distribution of ascidians in a semi-enclosed basin of the Mediterranean Sea. Journal of the Marine Biology Association of the United Kingdom, 88, 10531061.
  • Olden J.D., Hogan Z.S., Vander Zanden M.J. (2007) Small fish, big fish, red fish, blue fish: size biased extinction risk of the world's freshwater and marine fishes. Global Ecology Biogeography, 16, 694701.
  • Panetta P., Mastrototaro F., Matarrese A., Tanzarella S., Dalessandro S. (2004) Caratterizzazione biocenotica dei fondi mobili dei mari di Taranto: i molluschi. Biologia Marina Mediterranea, 11, 455460.
  • Parenzan P. (1969) Il Mar Piccolo e il Mar Grande di Taranto. Thalassia Salentina, 3, 1934.
  • Petronio B., Cardellicchio N., Calace N., Pietroletti M., Pietrantonio M., Caliandro L. (2012) Spatial and temporal heavy metal concentration (Cu, Pb, Zn, Hg, Fe, Mn, Hg) in sediments of the Mar Piccolo in Taranto (Ionian Sea, Italy). Water, Air, & Soil Pollution, 223: 863875.
  • Pierpaoli I. (1923) Prima contribuzione allo studio delle alghe nel Golfo di Taranto. Rivista di Biologia, 5, 693709.
  • Pierri C., Longo C., Giangrande A. (2010) Variability of fouling communities in the Mar Piccolo of Taranto (Northern Ionian Sea, Mediterranean Sea). Journal of the Marine Biology Association of the United Kingdom, 89, 19.
  • Prato E., Biandolino F. (2003) Seasonal changes in population of the Amphipod Gammarus aequicauda (Martynov, 1931). Mediterranean Marine Science, 4, 4956.
  • Salin K.R., Yohannan T.M., Mohanakumaran M. (2005) Fisheries and trade of seahorses, Hippocampus spp, in southern India. Fisheries Management & Ecology, 12, 269273.
  • Samoilys M.A. (1997) Manual for Assessing Fish Stocks on Pacific Coral Reefs. Queensland Department of Primary Industries, Brisbane.
  • Scalera-Liaci L., Corriero G. (1993) Distribution of the sponge fauna from the Mar Piccolo and the Mar Grande (Taranto, Ionian Sea). Biologia Marina, Supplemento al Notiziario SIBM, 1, 317318.
  • Sundin J., Jacobsson O., Berglund A., Rosenqvist G. (2011) Straight-nosed pipefish Nerophis ophidion and broad-nosed pipefish Syngnathus typhle avoid eelgrass overgrown with filamentous algae. Journal of Fish Biology, 78, 18551860.
  • Underwood A.J. (1997) Experiments in Ecology: Their Logical Design and Interpretation Using Analysis of Variance. Cambridge University Press, Cambridge: 504.
  • Vincent A.C.J. (1996) The International Trade in Seahorses. TRAFFIC International, Cambridge.
  • Vincent A.C.J., Hall H.J. (1996) The threatened status of marine fishes. Trends in Ecology & Evolution, 11, 360361.
  • Vincent A.C.J., Foster S.J., Koldewey H.J. (2011) Conservation and management of seahorses and other Syngnathidae. Journal of Fish Biology, 78, 16811724.
  • Wennhage H., Pihl L. (2007) From flatfish to sticklebacks: assemblage structure of epibenthic fauna in relation to macroalgal blooms. Marine Ecology Progress Series, 335, 187198.
  • Woodall L. (2009) Population genetics and mating systems of European seahorses Hippocampus guttulatus and Hippocampus hippocampus. PhD dissertationEcology, Evolution and Behaviour Group School of Biological Sciences Royal Holloway, University of London, London: 1425.
  • Woodall L.C., Koldewey H.J., Shaw P.W. (2011) Historical and contemporary population genetic connectivity of the European short-snouted seahorse Hippocampus hippocampus and implications for management. Journal of Fish Biology, 78, 17381756.
  • World Conservation Union. (2002). Available at
  • Zhang N., Xu B., Mou C.Y. (2003) Molecular profile of the unique species of traditional Chinese medicine, Chinese seahorse (Hippocampus kuda Bleeker). FEBS Letters, 550, 124134.