© 2004 by ICES/CIEM International Council for the Exploration of the Sea/Conseil International pour l'Exploration de la Mer
Quality of Nephrops as food for Atlantic cod (Gadus morhua L.) with possible implications for fisheries management
a Marine Research Institute Skúlagata 4, PO Box 1390, 121 Reykjavík, Iceland
b Department of Biology, University of Iceland Grensásvegur 12, 108 Reykjavík, Iceland
*Correspondence to B. Björnsson: tel: + 354 552 0240; fax: + 354 562 3790. e-mail: bjornb{at}hafro.is.
Nephrops was found to be of low quality as food for cod. In a laboratory experiment the mean specific growth rate of 1 kg cod was 0.184 and 0.415% d–1 when fed to satiation on Nephrops and capelin, respectively. This large difference in growth rate resulted not only from less intake of Nephrops (1.19 kg cod–1) than capelin (1.55 kg cod–1) but also because more Nephrops (4.6 kg) than capelin (2.2 kg) were required to produce each kilogramme of cod. Higher food conversion ratio was consistent with lower fat content of Nephrops (1.3%) than capelin (9.2%) but the exoskeleton also reduced the digestion rate of Nephrops. In the groups where Nephrops and capelin of equal mean weight were offered simultaneously, 40% of the diet consisted of Nephrops during the first week and 10% during the final seven weeks of the experiment. At the end of the experiment, condition factor, liver index, and gonadosomatic index were significantly lower for cod fed on Nephrops (0.967, 5.7, 7.1, respectively) than for those fed on capelin (1.086, 15.8, 11.2, respectively). These results suggests that predation by cod on Nephrops might be reduced by regular release of capelin or other similar food in the distributional areas of Nephrops.
Keywords: cod, food quality, Nephrops, predator–prey, stock enhancement
Received 11 March 2003; accepted 1 June 2004.
 |
Introduction |
|---|
 Top Introduction MethodsResultsDiscussionReferences |
|---|
Norway lobster (Nephrops norvegicus (L.)) is widespread in the northeast Atlantic, along the coast of western Europe and as far south as the Mediterranean Sea. The largest catches of Nephrops are obtained off the coasts of the British Isles and in the Bay of Biscay. The distribution of the Icelandic Nephrops stock is limited to the warmer waters off the south and southwest coast, at depths of 100–300 m and temperatures of 6–9°C (Eiríksson, 1999). The habitat of Nephrops is a soft bottom of clay, silt, or sand, where they dig extensive tunnels (Rice and Chapman, 1971) and feed on various invertebrates (Parslow-Williams et al., 2002) and discarded fish (Hrafnkell Eiríksson, Marine Research Institute, Reykjavík, personal communication). Cod (Gadus morhua L.) is thought to be the main predator of the Nephrops stocks living north of the English Channel. Studies of the diet of almost all the potential predators indicated that 88% of the predation on Nephrops in the Irish Sea was caused by cod (Symonds and Elson, 1983; Brander and Bennet, 1986). Qualitative analysis of stomach contents of cod and other fish species captured as by-catch in the Icelandic Nephrops fishery indicates that cod is by far the main predator of the Icelandic Nephrops stock (Hrafnkell Eiríksson, Marine Research Institute, Reykjavík, personal communication). On a yearly basis about 30% of the stomach content (by weight) of 50–90 cm cod from the Nephrops grounds in Icelandic waters consisted of Nephrops, making it the single most common prey item of cod in this area (Dombaxe, 2002).
As the price of Nephrops is much higher than the price of cod it is not desirable to let cod feed extensively on Nephrops. Two management strategies have been suggested to reduce predation on Nephrops. One is to increase the fishing mortality of the main predators locally in the Nephrops areas (Brander and Bennet, 1986, 1989). The other one is to offer the main predators in the Nephrops areas large quantities of alternative food, e.g. capelin (Mallotus villosus (Müller)), herring (Clupea harengus L.), blue whiting (Micromesistius poutassou (Risso)), or some other inexpensive fish (Björnsson, 2001).
The latter management strategy would not be feasible if cod would much rather eat Nephrops than the alternative food offered or if cod would grow much faster on Nephrops than the alternative food. In an attempt to answer these questions a laboratory experiment was carried out with the following aims: i) to estimate the relative food preference of cod for Nephrops and capelin of equal size and ii) to estimate the growth performance, nutritional condition, and feed conversion ratio of cod fed on either Nephrops or capelin or both.
|
Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
The experiment was carried out indoors at the Mariculture Laboratory, the Marine Research Institute Branch at Grindavík, southwest Iceland, using circular fibreglass tanks, 0.8 m deep with a diameter of 2.9 m. The seawater supply obtained from a 50-m-deep well had constant temperature and salinity (32). Temperature was measured in the tanks three times a week and the oxygen concentration weekly. The mean temperature during the experiment was 7.6°C. Water exchange was regulated so that the oxygen concentration in the outlet of each tank was above 7 mg l–1. Natural photoperiod (64°N) was maintained in all tanks with light intensity at the surface of about 25 lux. The experimental fish were obtained locally from the Grindavík laboratory where they had been hatched and raised.
Six tanks were used with 47 two and a half year old cod in each (the number of fish after the initial sampling). The mean initial weight was 989 g (range: 539–1781 g) and mean initial length was 43.8 cm (range: 34.7–53.0 cm). The fish were individually tagged with T-bar anchor tags (Floy tag and Manufacturing, Inc.). Two groups, Tanks 1 and 6 were fed on whole Nephrops only (Treatment 1), two groups (Tanks 2 and 5) on both capelin and Nephrops (Treatment 2), and two groups (Tanks 3 and 4) on capelin only (Treatment 3).
One batch of each prey was obtained for the whole experiment, the Nephrops from a research vessel in May 2000 and capelin from a commercial seiner in March 2000, both prey types were carefully packed in small boxes and frozen. The mean weight of the prey was 17.2 g ± 4.8 s.d. and 16.6 g ± 2.4 s.d. for the Nephrops and capelin, respectively. In the proximate analysis of the prey which was carried out by the Icelandic Fisheries Laboratories, Reykjavík, about 50 Nephrops and 50 capelin were ground up, homogenized and each measurement made twice and the average used. The analysis showed similar water and protein content but much lower fat content and higher ash content in the Nephrops compared to the capelin samples (Table 1). Calculations showed that gross energy content per g wet weight of Nephrops was much lower than that of capelin.
|
The experimental fish had only received commercial dry feed until they were tagged and weaned to the experimental feed 1 week prior to the experiment. The experiment lasted for a period of 4 months. The fish were individually weighed and their length measured three times, at the start (28 September 2000), after 2 months (30 November 2000), and at the end of the experiment (5 February 2001) to estimate their growth rates. The fish were not fed 3 days before weighing. The fish were handfed to satiation three times a week. The uneaten food was removed and weighed to accurately determine the actual food intake. The food preference was determined in Tanks 2 and 5 where both Nephrops and capelin were offered simultaneously. When one type of feed was finished more of that type was added until a few food items of both types remained.
To estimate their initial condition 17 fish were sampled randomly, 2 or 3 fish from each tank, and eviscerated to measure gutted weight, weight of liver, gonads, and stomach content. The procedure was repeated 2 months later, with five fish randomly sampled from each tank. At the same time (30 November) all abnormal fish, i.e. fish which were very skinny or with popeye were removed from the experiment, four fish from Tank 1, three from Tank 6, and one from Tank 2. A total of five fish died in the experiment, two in Tank 1, one in Tank 3, and two in Tank 4. At the end of the experiment all the remaining fish were eviscerated and measured as above. The following indices of nutritional status and growth rate were used: condition factor (CF), liver index (LI), gonadosomatic index (GSI), and specific growth rate (SGR), CF = 100 gutted weight length–3, LI = 100 weight of liver (gutted weight)–1, GSI = 100 weight of gonads (gutted weight)–1, SGR = 100(ln W2 – ln W1)t–1, where W1 and W2 are the initial and final weights and t the time in days in each growth period. The food conversion ratio (FCR) of each group was calculated as the total food intake divided by the total weight gain of the fish during the growth period.
One-way nested ANOVA (Zar, 1996) was used to determine differences within and among treatments as each tank was assigned to one treatment only. Tukey multiple comparison procedure was used to identify the differences in the treatments means in all the parameters of interest.
|
Results |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Diet had a large effect on growth rate (
L, W, SGR) and nutritional condition (CF3, LI3, GSI3) but there was no statistical difference between replicates, i.e. Tank 1 vs. Tank 6, Tank 2 vs. Tank 5, and Tank 3 vs. Tank 4 (Table 2). It was found with multiple comparisons that the groups receiving only Nephrops grew slower and were in worse nutritional condition than the groups receiving both Nephrops and capelin and the ones receiving only capelin. For all the variables above there was a highly significant difference (p < 0.001) between Treatment 1 and 2, Treatment 1 and 3, but not between Treatment 2 and 3 (p > 0.05), except for LI3 which was also significantly higher in Treatment 3 than in Treatment 2.
|
The average length gain was much less in Treatment 1 (3.6 cm) than in Treatment 2 (5.2 cm) and Treatment 3 (5.2 cm) (Table 3). The average weight gain was only 295 g in Treatment 1 compared to 634 and 724 g in Treatments 2 and 3, respectively. Treatments 2 and 3 were not significantly different (p > 0.05) from each other, but they both differed significantly from Treatment 1 (Table 3, Figure 1).
|
(Nephrops), Treatment 2 
(Nephrops and capelin), and Treatment 3 
(capelin).
|
Mean SGR was much lower in Treatment 1 (0.184% d–1) than in Treatment 2 (0.396% d–1) and Treatment 3 (0.415% d–1) (Table 3, Figure 2). The maximum growth rate of the fish receiving only Nephrops (0.367% d–1) was much lower than the maximum receiving both Nephrops and capelin (0.657% d–1) or only capelin (0.641% d–1) (Figure 2). There was less difference in the minimum growth rate and some fish in all groups grew little or lost weight during the experiment (Figure 2).
|
The total food consumption was 100.6 kg Nephrops in Treatment 1, 27.2 kg Nephrops (19%) and 112.9 kg capelin (81%) in Treatment 2, and 135.5 kg capelin in Treatment 3. For the whole experiment, the mean food consumption per fish was 1.19, 1.58, and 1.55 kg and the average feed conversion ratio (FCR) 4.6, 2.5, and 2.2 in Treatments 1, 2, and 3, respectively (Table 3). In both growth periods the FCR was much higher and the mean SGR much lower in Treatment 1 than in Treatments 2 and 3 (Table 4) but there was not a significant difference between Treatment 2 and 3 (Tukey multiple comparison). During the first week of the experiment the groups receiving both prey types of equal mean weight ate 40% Nephrops and 60% capelin. During the 12th week the diet consisted of 10% Nephrops and 90% capelin, this ratio remaining unchanged till the end of the experiment (Figure 3).
|
|
At the conclusion of the experiment, the average condition factor (CF) in Treatment 1 (0.967) was significantly lower than in Treatment 2 (1.073) and Treatment 3 (1.086) (Table 5). However, the overlap was considerable (Figure 4). The final liver index (LI) was only 5.7 in Treatment 1, 13.1 in Treatment 2, and 15.8 in Treatment 3. In this case there was also a significant difference between Treatment 2 and 3 (Table 5, Figure 5). During the experimental period LI decreased by 58% in Treatment 1 and 4% in Treatment 2, but increased by 15% in Treatment 3 (Table 5), showing how sensitive LI of cod is to the energy content of the prey (Table 1).
|
|
|
At the end of the experiment, mean gonadosomatic index (GSI) was lower in Treatment 1 (7.1) compared to Treatment 2 (11.7) and Treatment 3 (11.2) (Table 5), 87% of the fish in Treatment 1 being mature compared to 95% in Treatment 2 and 95% in Treatment 3 (Table 6). In all treatments there was a large variation in GSI between individual fish and great overlap between treatments, although the maximum GSI was markedly lower in Treatment 1 (18.8) than in Treatment 2 (29.4) and Treatment 3 (26.2). Mean GSI was much higher for the males, 10.0, 16.2, and 16.1, than females, 4.3, 6.3, 6.5, for Treatments 1, 2, and 3, respectively (Table 6).
|
At the termination of the experiment, food remains were only found in 1 cod in Treatment 3 but in 46 cod in Treatment 1 (mean stomach content 0.02 and 6.6 g fish–1, respectively), showing clearly that cod digest Nephrops much more slowly than capelin of equal mean weight.
|
Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
The average growth rate of cod fed to satiation on Nephrops was less than half of those fed on capelin or on a mixed diet of capelin and Nephrops. The growth rate of cod receiving capelin and Nephrops (Treatment 2) and only capelin (Treatment 3) was similar to that predicted by the growth model of Björnsson and Steinarsson (2002), 106% and 111% of the value given by the model in Treatments 2 and 3, respectively. The low mean growth rate of cod in Treatment 1 suggests that Nephrops is of low quality as food for cod. However, there was a large individual variation in growth rate of cod in all treatments, a few fish even losing weight. Much of this variation may have been due to social interaction. Therefore it is also useful to compare the fastest growing fish in each group. The results show clearly that the maximum growth rates of the fish being fed solely on Nephrops are much lower than those being fed on capelin or a mixture of the two prey types. Reduced maximum growth rate of cod eating only Nephrops may result from less food intake, lower energy content, and lower digestability of Nephrops than capelin, the reference food.
The food intake per cod in Treatment 1 (Nephrops) was 25% less than in Treatment 2 (capelin and Nephrops) and 23% less than in Treatment 3 (capelin). The rigid exoskeleton with numerous appendages makes it impossible for cod to pack their stomachs as tightly as with softer and more flexible capelin (Bromley, 1991). However, the packing problem of Nephrops becomes less when some capelin is also consumed. Less palatability of Nephrops compared to capelin may also have affected the total food consumption. The cod used in the experiment had never seen Nephrops or capelin but both feed types were readily accepted. However, the fish had more difficulty in swallowing Nephrops than capelin of similar weight. The cod were often seen swimming for some time with Nephrops in the mouth before swallowing or spitting it out. In the groups where Nephrops and capelin of equal mean weight were offered simultaneously, 40% of the diet consisted of Nephrops during the first week and only 10% during the seven final weeks. Even though capelin is clearly preferred it seems that cod still want some variety in the diet.
The energy content of Nephrops was much less than that of capelin. The total protein content of the experimental diet was slightly higher for Nephrops (14.3%) than capelin (13.5%) but when considering the amount of chitin in the exoskeleton it is likely that digestible protein content was less in the Nephrops than in the capelin. The exoskeleton of Nephrops consists of about 70% calcium salts, 21% chitin, and 9% proteins (Welinder, 1974). The fat content of Nephrops (1.3%) was much lower than that of capelin (9.2%) which means that for cod eating only Nephrops most of the energy for metabolism must come from protein instead of fat which will result in low growth efficiency (Brett and Groves, 1979) and thus a high feed conversion ratio.
The thick and hard exoskeleton reduces the rate of digestion of Nephrops. While virtually all the capelin had been evacuated a substantial part of the Nephrops still remained in the stomach after 72 h. Bromley (1991) found that Nephrops was evacuated much slower than sprat (Sprattus sprattus) of similar size. The daily gastric evacuation rate of Nephrops (0.88% of body weight d–1) determined for 790 g cod at 8.8°C by Bromley (1991) was similar as the average food consumption (FC) of Nephrops (FC = SGR x FCR = 0.184 x 4.6 = 0.85% of body weight d–1) by 1174 g cod at 7.6°C in the present experiment. The evacuation rate of 2–4 g prawn (Pandalus borealis Krøyer) has been found to be slower compared to that of 4–48 g herring and 16 g capelin (dos Santos and Jobling, 1992). This would appear to put cod feeding predominately on crustaceans with thick and hard exoskeleton at a disadvantage.
The lower food intake, lower energy value, and slower gastric evacuation rate of Nephrops compared to capelin was also reflected in other indices of nutritional condition. At the end of the experiment, the mean liver index was almost three times higher for cod offered capelin instead of Nephrops. Most of the fish eating only Nephrops had low liver index (<10%) resembling the wild, inshore cod on the east coast of Iceland (Björnsson, 2002) which feed mostly on invertebrates (Björnsson, 1999b). Most of the fish feeding on capelin had high liver index (>10%) whereas most of the fish feeding on the mixed diet had liver index ranging between 6% and 18%. Although the mean gonadosomatic index (GSI) and condition factor (CF) were significantly lower for Treatment 1 compared to Treatments 2 and 3 the difference was much less than for the liver index and growth rate, suggesting that liver index is a more sensitive indicator of energy intake and growth rate than GSI and CF. A useful comparison of liver index and CF for cod were made by Lambert and Dutil (1997).
On a yearly basis Nephrops was the most common prey (
30% by weight) of 50–90 cm cod on the Nephrops grounds off the south coast of Iceland (Dombaxe, 2002). The rest of the diet consisted almost entirely of fish, mostly blue whiting, Norway pout (Trisopterus esmarki (Nilsson)), eelpot (Lycodes esmarki Collett), haddock (Melanogrammus aeglifinus (L.)), capelin, long rough dab (Hippoglossoides platessoides limandoides (Bloch)), and whiting (Merlangius merlangus merlangus L.). However, there were large seasonal changes in the relative contribution of Nephrops in the diet, from 10% in March during the spawning migration of capelin to 65% in July when few fish prey were available (Dombaxe, 2002). Usually the fish part of the diet was more than 50% of the total diet. Thus, the diet appears to have been nutritious and well balanced and no reason to believe that food quality as such limited the growth rate of these fish. However, the mean length of 3- and 5-year-old cod sampled in the annual Nephrops survey in May 2000 and May 2001 was 46.0 and 65.5 cm, respectively (Dombaxe, 2002), suggesting a growth rate of 4-year-old cod as 9.8 cm yr–1 which is similar to that found in the Icelandic ground fish survey (Björnsson, 1999b). This is much lower growth rate than the food-unlimited growth rate found in sea pens (Björnsson, 1999b), laboratory experiments (Björnsson et al., 2001; Björnsson and Steinarsson, 2002), and experimental feeding of wild free-ranging cod in an Icelandic fjord (Björnsson, 1999a, 2002) indicating that the growth rate of these fish were food limited.
As Nephrops is 3–4 times more expensive than cod and 4.6 kg of Nephrops are required to produce each kilogramme of cod it is clearly undesirable from an overall management point of view that cod feed and grow on Nephrops (Björnsson, 2001). About 50% of the Nephrops consumed by cod were larger than the fishable size (>35 mm CL) (Dombaxe, 2002). Any reduction in the predation on Nephrops in this size category could increase the total allowable catch (TAC) of the Nephrops fishery. Furthermore, reduced predation on smaller Nephrops would most likely increase the recruitment to the fishable part of the stock.
Two management strategies to reduce predation on Nephrops have been suggested. Brander and Bennet (1989) proposed to increase the fishing mortality of cod locally in the Nephrops areas, whereas Björnsson (2001) suggested to offer cod alternative food in or near the Nephrops areas. The evidence that cod prefer capelin to Nephrops and that cod on the Nephrops grounds are food-limited and prey heavily on Nephrops suggests that large-scale release of capelin, blue whiting, or other fish prey is likely to attract cod to the feeding areas. As long as sufficient amount of suitable feed is provided the predation by cod on the Nephrops stock will be reduced. Large numbers of predators in the feeding areas are, however, potential threat to the Nephrops stock as they may switch to preying on Nephrops if not enough feed is provided regularly. Therefore, harvesting of cod or other potential predators must be carried out periodically in the feeding areas to further reduce the risk of predation. Thus, largest yields of Nephrops may be obtained by a combination of both management strategies. However, these strategies do not imply that the cod stock as a whole is to be exploited more heavily, instead they imply that more of the cod quota is to be taken in or near the Nephrops areas where anthropogenic feeding takes place and less of it in other areas.
|
Acknowledgements |
|---|
We thank the staff at the Mariculture Laboratory for taking good care of the fish during the experiments and also thank Hrafnkell Eiríksson and Dr Karl Gunnarsson at the Marine Research Institute (MRI) for their critical reading of the manuscript. Dr Andrzej Jaworski, MRI helped with the statistical analysis of the data and two anonymous referees made useful comments concerning the manuscript.
|
References |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
-
Björnsson B. (1999) Fjord-ranching of wild cod in an Icelandic fjord: effects of feeding on nutritional condition, growth rate, and behaviour. In Howell B.R., Moksness E., Svåsand T. (Eds.). Stock Enhancement and Sea Ranching(Fishing News Books, Blackwell Science, Oxford) pp. 243–256 606 pp.
Björnsson B. (1999) Is the growth rate of Icelandic cod (Gadus morhua L.) food-limited? A comparison between pen-reared cod and wild cod living under similar thermal conditions. Rit Fiskideildar 16:271–279.
Björnsson B. (2001) Can fisheries yield be enhanced by large-scale feeding of a predatory fish stock? A case study of the Icelandic cod stock. Canadian Journal of Fisheries and Aquatic Sciences 58:2091–2104.
Björnsson B. (2002) Effects of anthropogenic feeding on the growth rate, nutritional status and migratory behaviour of free-ranging cod in an Icelandic fjord. ICES Journal of Marine Science 59:1248–1255.
Björnsson B. and Steinarsson A. (2002) The food-unlimited growth rate of Atlantic cod (Gadus morhua). Canadian Journal of Fisheries and Aquatic Sciences 59:494–502.
Björnsson B., Steinarsson A., Oddgeirsson M. (2001) Optimal temperature for growth and feed conversion of immature cod (Gadus morhua L.). ICES Journal of Marine Science 58:29–38.
Brander K.M. and Bennet D.B. (1986) Interactions between Norway lobsters (Nephrops norvegica L.) and cod (Gadus morhua L.) and their fisheries in the Irish Sea. Canadian Special Publication of Fisheries and Aquatic Sciences 92:269–281.
Brander K.M. and Bennet D.B. (1989) Norway lobsters in the Irish Sea: Modeling one component of a multispecies resource. In Caddy J.F. (Ed.). Marine Invertebrate Fisheries: Their Assessment and Management(John Wiley and Sons, New York) pp. 183–204 752 pp.
Brett J.R. and Groves T.D.D. (1979) Fish Physiology. In Hoar W.S., Randall D.J., Brett J.R. (Eds.). Bioenergetics and Growth(Academic Press, New York) vol. VIII: pp. 279–352 786 pp.
Bromley P.J. (1991) Gastric evacuation in cod (Gadus morhua L.). ICES Marine Science Symposia 193:93–98.
Dombaxe M. Á. D. (2002) Predation intensity on Norway lobster (Nephrops norvegicus) by cod (Gadus morhua), in Háfadjúp, southwest Iceland and the relative prey preference of cod fed on capelin (Mallotus villosus) and Norway lobster. M.Sc. thesis, University of Iceland. 84 pp.
Eiríksson H. (1999) Spatial variabilities of CPUE and mean size as possible criteria for unit stock demarcations in analytical assessments of Nephrops at Iceland. Rit Fiskideildar 16:239–245.
Lambert Y. and Dutil J-D. (1997) Can simple condition indices be used to monitor and quantify seasonal changes in the energy reserves of Atlantic cod (Gadus morhua)? Canadian Journal of Fisheries and Aquatic Sciences 54:Suppl. 1, 104–112.
Parslow-Williams P., Goodheir C., Atkinson R.J.A., Taylor A.C. (2002) Feeding energetics of the Norway lobster, Nephrops norvegicus in the Firth of Clyde, Scotland. Ophelia 56:101–120.[Web of Science]
Rice A.L. and Chapman C.J. (1971) Observations on the burrows and burrowing behaviour of two mud-dwelling decapod crustaceans, Nephrops norvegicus and Goneplax rhomboides. Marine Biology 10:330–342.[CrossRef]
dos Santos J. and Jobling M. (1992) A model to describe gastric evacuation in cod (Gadus morhua L.) fed natural prey. ICES Journal of Marine Science 49:145–154.
Schmidt-Nielsen K. (1979) Animal Physiology: Adaptation and Environment(Cambridge University Press, Cambridge) 560 pp.
Symonds D. J. and Elson J. M. (1983) The food of selected fish species on Nephrops grounds in the western Irish Sea. ICES, CM 1983/K: 8. 14 pp.
Welinder B.S. (1974) The crustacean cuticle – I. Studies on the composition of the cuticle. Comparative Biochemistry and Physiology 47A:779–787.[CrossRef][Medline]
Zar J.H. (1996) Biostatistical Analysis(Prentice Hall, New Jersey) 662 pp.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||