Feeding habits of stargazer ( Uranoscopus scaber Linnaeus, 1758) in the southern Adriatic Sea (Croatia) Prehrana bežmeka (Uranoscopus scaber Linnaeus, 1758) u

* The feeding habits of the stargazer Uranoscopus scaber Linnaeus, 1758 were analysed using 360 stomachs of specimens collected in southern Adriatic Sea between October 2011 and September 2012. The total body length of the specimens ranged from 9.7 to 32.1 cm (18.89 ± 3.34 cm) and weight from 11.7 to 618.7 g (112.79 ± 63.17 g). In the sampled population there were 213 females (59.17%) and 147 males (40.83%). Stargazer is carnivorous fi sh and its diet in Adriatic Sea was composed of teleost fi sh, cephalopods and crustaceans. According to numerical abundance fi sh predominated followed by crustaceans, while according


INTRODUCTION
Members of Uranoscopidae family are benthic predators that live at depths of up to 700 meters, and can grow up to 75 cm in length and 11 kg in weight [1].The main characteristics of the family are an elongated and strong, laterally fl attened body and a massive, square and bony head.Their eyes are located on the top of their heads while their mouths are vertical with a protractile appendix attached to the mandible.The teeth are small [1,2].The stargazer, Uranoscopus scaber Linnaeus, 1758 is the only representative of the Uranoscopidae family in Mediterranean Sea and eastern Atlantic from the Bay of Biscay to Senegal [2].In the Adriatic Sea it is recorded throughout the entire area except in the deep sea and shallow coves of the northern part [3].The maximum reported total length of stargazer from the Adriatic Sea is 36 cm.The only larger specimen (TL=38 cm) in the entire area of its natural distribution was found in the records of the International Game Fishing Association [4].The reproduction period of the stargazer ranges from March to September [5] and the length at fi rst maturity is 11.76 cm for males 13.75 cm for females [6].Although stargazer is not a commercially important species in fi sheries, it is an important component of the food chain [5,7,8,9].This benthic predatory species burrows into the muddy-sandy bottom and lure prey.When lunges out of the substrate it bends body between the head and trunk by more than 60° [1,10].Stargazer does not use the electrical impulses to capture prey, but releases them during mechanical stimulation.The tissue that releases electrical impulses has lost the properties of ordinary muscle tissue, but still does not have the properties of electrical organs [11].The diet of the stargazer is known from the few studies reported from diff erent parts of the Mediterranean [7,12,13].For the Adriatic Sea, only Jardas [2] noted that it feeds on fi sh, polychaetes and crustaceans.Information on diet composition of commercial and noncommercial species is crucial for implementing a multi-species approach to fi sheries management [14] so this study aims to determine, for the fi rst time, the diet and feeding habits of the stargazer from Adriatic Sea, by analysing variations in diet composition, taking into account factors such as seasonality and size groups.

MATERIAL AND METHODS
The 360 individuals of stargazer analysed in this study were collected monthly from October 2011 to September 2012 in the southern Adriatic Sea (Figure 1).This area is characterized by a seabed covered with muddy sediment [2].The specimens were collected during day time hours at depths of 100 to 140 m using a commercial bottom trawl net.The net was 45 m long with 24 mm stretched mesh cod-end.The duration of each haul was about 3 h and the trawling speed fl uctuated from 2.6 to 2.9 knots.After the catch the samples were stored on ice and transferred to the laboratory.There was no evidence of regurgitation.Total length (TL) of specimens was measured to the nearest 0.1 cm and weight (W) to the nearest 0.1 g per individual.Sex of fi sh was determined by eye examination of gonadal tissue.Stomachs were dissected and weighed to the nearest 0.001 g.The stomach contents and each individual prey Figure 1 Sampling location in the southern Adriatic Sea (Croatia) were also weighed.When possible, prey was determined to the lowest possible taxon depending on condition.The analysis of changes in feeding habits was expressed as the vacuity index (%V) (%V = ratio of the number of empty stomachs x 100 and a total number of stomachs) and the fullness index (% Jr) (% Jr = (Wp/W) x100 where Wp is the mass of prey items calculated as the diff erence between the mass of an intact stomach and an empty stomach and W is total body mass) [15].Student t-test was used to test the signifi cance of the diff erence of vacuity index values between seasons [16].Diet breadth was calculated oversize and season using the Shannon-Wiener diversity index: H'= -Σ pi x ln pi, where pi is the proportion of individuals belonging to a given species [17].
Three indices were used to describe diet composition: percentage frequency of occurrence (%F = ratio of stomachs that contained a given prey and the number of total non-empty stomachs x 100), percentage numerical abundance (%N = ratio of prey in a given taxonomic group and to the total number of prey in all groups x 100), and percentage gravimetric composition (%W = ratio of total weight of a particular taxonomic group and the total weight of prey in all groups found x 100) [15,18,19].
For more results about diet index of relative importance (IRI = (%N + %W) + %F), the main food index (MFI = [(%N + %F)/2] x W) and the coeffi cient of nutritiveness (Q = %N x %W) were calculated [19,20,21] for the whole sample, for each season separately and by sex.

RESULTS
Total body length of 360 individuals ranged from 9.7 to 32.1 cm (18.89 ± 3.34 cm) and weight from 11.7 to 618.7 g (112.79 ± 63.17 g).The sample was composed of 213 females (59.17%) and 147 males (40.83%).The total body length of females ranged from 12.8 to 32.1 cm (19.5 ± 2.93 cm) and that of males from 9.7 to 27.5 cm (18 ± 3.7 cm) (Figure 2).Of the 360 stomachs of stargazers 40 were completely empty so the vacuity index was relatively low (%V = 11.11%).Food that could be determined was found in 126 stomachs (35%), while completely digested gut content was found in 194 stomachs (53.89%).As a result the value of the fullness index of stargazer was relatively low (% Jr = 1.50%).Changes in the annual vacuity index were statistically signifi cant (t = 4.90, P = 0.00).The highest number of empty stomachs and the highest index were found in winter (%V = 15.55%) and the lowest in spring (%V = 7.78%) (Table 1).For each season 90 stomachs were analysed.The diet of stargazer was also analysed according to diff erent size groups.All individuals were divided into four total length groups: I <15 cm (n = 46), II = 15-20 cm (n = 170), III = 20-25 cm (n = 136) and VI >25 cm (n = 8).The highest vacuity index was recorded for size group II (% V = 5.83) and the lowest for group VI (Table 1  The diet of stargazers in the Adriatic Sea consisted of three main taxonomic groups: teleost fi sh, crustaceans and cephalopods (Table 2).By abundance and weight, the most important food was teleost fi sh (%N = 82.27;%W = 63.42), while cephalopods (%N = 14.58; %W = 17.42) and crustaceans (%N = 10.42;%W = 14.13) were additional food.The diff erent groups of prey indicate a relatively diverse diet of the stargazer in the southern Adriatic Teleost fi sh were the important and main food of stargazer in the southern Adriatic (IRI = 167.1;MFI = 508.02;Q = 805.173). A. sphyraena was the dominant food in summer, while M. merluccius was the dominant food in autumn, winter, and spring.Compared to the rest of the year during spring season abundance of cephalopods in diet decreased (IRI = 6.45), while for crustaceans increased (IRI = 66.41).
Variation in diet composition was analysed in relation to the total body length.Evident is that the teleost fi sh dominated in diet in fi rst three size groups.A. sphyraena was the main food for size group I (IRI = 66.24) and II (IRI = 38.32)while M. merluccius was for the size group III (IRI = 34.43)(Table 4).Both species were recorded in the stomachs of stargazer whose total body length is less than 25 cm.In size group IV dominant prey was P. longirostris (IRI = 62.21).M. merluccius was also recorded but A. sphyraena was not.

DISCUSSION
The stargazer is carnivorous fi sh and although three groups of prey: teleost fi sh, crustaceans and cephalopods were recorded, a signifi cant dominance of teleost fi sh in the stomachs was confi rmed.This can be explained by the state of prey populations on the seabed in the sampling area, which is confi rmed by commercial fi shing catches and indirectly by examining the stomachs of the researched species.The dominant species in the stomachs were A. sphyraena (IRI = 41.47) and M. merluccius (IRI = 38.75).Booth species in the Adriatic Sea generally lives at depths between 100 and 200 m and during recruitment the highest density of booth species are found at depths of 100 to 150 m where the stargazer in this research was sampled.However, Sanz [12] in his investigation on diet reported M. merluccius as prey for stargazer in the Balearic Sea, but not as a primary food.Principal pray were fi sh, primarily Callioymus maculatus and Trachurus trachurus.C. maculatus was also recorded in this study, but with a lower percentage of occurrence.In the Egyptian Mediterranean waters, the dominant prey recorded in the stomachs of stargazer was also fi sh, Spicara smaris [13], which was not the case in our research.This can be explained also by the depth during sampling because S. smaris in the Adriatic Sea usually lives at depths of 10 to 40 m [2].Species recorded in the diet of the stargazer in the Black Sea [7] Merlangius merlangus and Gobius sp. were also recorded in this study, but with a lower percentage of occurrence.
Loligo vulgaris the most numerous and important cephalopod species in this study (IRI = 22.58) was also the only cephalopod species recorded in the stomachs of stargazers from the Balearic Sea [12].The most abundant and important crustacean species was Solenocera membranacea (IRI = 18.31) and in the Adriatic Sea, the highest abundance of this species was found at depths between 50 and 100 m [22].In the stomachs of stargazers from the Balearic Sea this species was the second most abundant crustacean [12].The abundance of species on which the stargazer feeds depends primarily on locality.The availability of each species also depends on its seasonal occurrence.According to the highest IRI values, A. sphyraena was the primary food for stargazer in the Adriatic Sea in summer while M. merluccius was primary food throughout the rest of the year.M. merluccius spawns almost all year round with peaks in winter and summer [23,24] suggesting an abundant food supply for stargazers.In relation to the total body length teleost fi sh dominated in diet in size group I, II and III.In group IV dominant prey was P. longirostris but this group had the smallest number of samples (n = 8), and this diff erence in important prey could be result of limited samples.The important prey in group I and II was A. sphyraena while in group III was M. merluccius.
The values of vacuity index for the whole sample was relatively low (%V = 11.11%)suggesting that the stargazer is an active feeder.The abundance of available prey in the environment may have a major infl uence on the lower values of empty stomachs of demersal fi shes from Adriatic Sea [25].A notable higher percentage of empty stomachs for the whole sample have been reported for the coast of Egypt (34.8%) [13] and the Black Sea (35.4%) [7].
In this research the changes in vacuity index were statistically signifi cant.The highest vacuity index was in winter and the The most diverse diet was recorded in the spring season (H´=3.65)and feeding on high number of prey items was recorded for the size group III (H´= 3.64).Stargazer is predator that feeds by bending the body trunk when lunges out of the substrate which causes the prey to be sucked in [10].
Feeding habits in all researched areas primarily depends on ecological conditions that will at certain moments enable the abundance of certain species of prey.The species that we found in the stomachs of stargazer tell us about the situation with the populations of organisms on the seabed of the researched area.As a predator who passively waits on the seabed and whose nutrition depends on the movement of its prey the qualitative and quantitative development of its population depends on the richness of the populations in the researched area.Seasonal fl uctuations of the population of prey as well as the presence throughout the year determine the dominant pray which in our research was teleost fi sh, mostly these two species, M. merluccius and A. sphyraena.Both of them spatially overlap with stargazer.The spawning season and the available amount of fry of these two species certainly favour the maintenance of the stargazer population in this water area, which is proven by the contents of the stomachs throughout the researched period.Such a conclusion certainly depends on the level of human activity in the researched area because increased fi shing eff ort can signifi cantly disrupt the relationships between the populations of organisms mentioned in the research.Certainly, these results show the current situation which may change in the future where the causes may be diff erent but human activity, as one of them, is certainly not negligible.

Table 1
).Values of Vacuity index (%V) and Shannon-Wiener diversity index (H´) of stargazer Uranoscopus scaber in the southern Adriatic Sea by seasons and total length groups

Table 2
Diet composition of stargazer Uranoscopus scaber in the southern Adriatic Sea (%F = frequency of occurrence, %N = numerical abundance, %W = gravimetric composition, IRI = index of relative importance, MFI = main food index, Q = coeffi cient of nutritiveness)

Table 3
Seasonal variation in diet composition of stargazer Uranoscopus scaber in the southern Adriatic Sea (IRI = index of relative importance, MFI = main food index, Q = coeffi cient of nutritiveness)

Table 4
[13]ation in diet composition in relation to the total body length of stargazer Uranoscopus scaber in the southern Adriatic Sea (IRI = index of relative importance) in spring which coincides with the spawning season from March to September with a peak in May[5].Feeding intensity was increased before spawning.Rizkalla and Philips[13]also recorded the highest value of vacuity index in winter and the lowest in spring.Related to size the highest vacuity index for the stargazer from Adriatic Sea was recorded in group II and the lowest in group VI (no empty stomachs were recorded). lowest