THEME 1
Environment, ecosystem biology, habitat and diversity, oceanography

Environmental and biological aspects of deepwater demersal fishes

J.D.M. Gordon
Scottish Association for Marine Science
Oban PA37 1QA, UK
<[email protected]>

1. INTRODUCTION

1.1 Deepwater fisheries

Some artisanal deepwater^[29] fisheries, such as the hook and line fisheries for black scabbardfish (Aphanopus carbo) in the Atlantic and Ruvettus spp. in the Pacific have a long history. The more recent development of highly mechanized and efficient deepwater fisheries targeting new species, such as macrourids (grenadiers), armourhead (Pseudopentaceros wheeleri) and orange roughy (Hoplostethus atlanticus) in the Atlantic and Pacific can be traced back to the exploratory fishing by vessels of the USSR in the 1960s (Gordon 2001a). In the Atlantic the grenadiers, mainly the roundnose grenadier (Coryphaenoides rupestris) but also roughhead (Macrourus berglax), were the main target species and landings peaked in the 1970s. The fishery for roundnose grenadier revived in the late 1980s when French vessels began to target this species, and also several other deepwater species, on the European continental margin. The armourhead fishery on Pacific seamounts began in the mid 1960s, first by the USSR and later by Japanese trawlers (Humphreys and Moffitt 1999). The South Pacific fishery for orange roughy was developed by New Zealand in the 1970s (Clark 2001) and later by Australia in the 1980s (Koslow et al. 1994).

The global trend of increasing catches of deepwater fish has been analysed by Garibaldi and Limongelli (2003) using the available FAO statistics. They extracted all catch data on oceanic species and further subdivided these data into epipelagic (tuna-like species, oceanic sharks, cephalopods and krill) and deepwater species (including shrimps and crabs). The deepwater catch was about 2 percent of the total oceanic catch until about 1975. Thereafter it increased to a level of about 20 percent and in the most recent years 1998 and 1999, reached 33 percent at over 2 million tonnes. Gadiform fishes, especially blue whiting (Micromesistius poutassou), dominated the deepwater group.

It was the discovery of commercial quantities of orange roughy in the eastern North Atlantic in the early 1990s that generated significant interest in deepwater fisheries in European waters. This resulted in a symposium, organized jointly by the Sea Fish Industry Authority and the Scottish Association for Marine Science (SAMS), on the Deepwater fisheries of the North Atlantic oceanic slope, held in 1994 (Hopper 1995). In the introductory paper entitled Environmental and biological aspects of slope-dwelling fishes of the North Atlantic (Gordon, Merrett and Haedrich 1995) we addressed the following questions. (1) How do the physical features of the continental slope and shelf compare? (2) How does the physical environment of the slope differ from that of the shelf? (3) How do the demersal species assemblages on the shelf and slope differ from one another and are the latter basically different from the pelagic oceanic assemblage? (4) How do the basic distribution patterns (vertical and horizontal) of slope dwellers compare with their shelf-dwelling counterparts? (5) How does the vertical distribution pattern of fish biomass correlate with the trophic input to the oceanic environment? and (6) What is known about deep-sea fish population structure and breeding biology? In this paper I revisit these questions, slightly modified to embrace a wider global scale, to assess how much our knowledge has advanced over the last decade. In this context the proceedings of several recent international meetings on deepwater fishes and/or fisheries are relevant. These were the 1996 Deepwater fishes symposium of the Fisheries Society of the British Isles (McIntyre and Thorpe 1996), the ICES 1998 Deepwater fish and fisheries theme session (Gordon 2001b) and the NAFO 2001 Deep-sea fisheries symposium (Moore and Gordon 2003).

Deepwater fisheries are generally considered to be those that exploit fish or shellfish that habitually live at depths greater than 400 m. However, this is an arbitrary boundary since many species have ranges that extend from the continental shelf into deepwater. Others, such as the sablefish (Anoplopoma fimbria) of the northern Pacific, occupy the shelf as juveniles and the deepwater as adults. In the northeast Atlantic species such as ling (Molva molva), tusk (Brosme brosme) and anglerfish (Lophius piscatorius) are generally considered as species of the continental shelf but all are also found on the slope, especially the anglerfish which can be found at 1 000 m depth. There are also fisheries on the upper continental slopes for species with close affinities to shelf species such as the Cape hakes (Merluccius capensis and M. paradoxus) off southern Africa, the blue whiting (Micromesistius) fisheries in the northern and southern hemispheres, the deepwater redfish (Sebastes spp.) fisheries of the Atlantic and Pacific and the Greenland halibut (Reinhardtius hippoglossoides) of the North Atlantic and Pacific. In this paper the emphasis will be on the ‘new’ deepwater fish species.

1.2 Terminology

On the continental shelf fish and fisheries are usually classified as pelagic or demersal. In the deep sea the pelagic extends from the surface to abyssal depths and it is usual to divide it into three zones. The epipelagic zone includes all those fish living in the upper photic layer of the ocean, such as the tuna fishes. The mesopelagic zone spans the depth range from below the photic zone down to about 1 000 m and supports an abundant and diverse fish fauna, such as myctophids (lanternfish) and gonostomids (bristle mouths). Many mesopelagic fishes and invertebrates are diel vertical migrators. They migrate to the surface to feed at night and return to the depths during the day and in doing so form an important link in the deepwater food-chain (see Section 6). Some mesopelagic lanternfishes that form dense aggregations are, or have been, exploited in areas such as off South Africa, in the Arabian Sea and the Southern Ocean (Gjøsaeter and Kawaguchi 1980). The bathypelagic fishes occur from about 1 000 m down to abyssal depths and are generally highly adapted, often in bizarre ways, to life in a dark, food-poor environment. They are low in abundance and biomass and are never likely to have any commercial value. Typical examples are the deepwater angler fishes (family Ceratidae) and the gulper eels (family Eurypharyngidae).

In the deep sea the demersal fishes are generally divided into two categories, benthic and benthopelagic. The benthic fishes are those that have a close association with the seabed and include species such as skates, flatfishes and tripod fishes (Bathypterois spp.). Benthopelagic fishes are those that swim freely and habitually near the ocean floor and, in the areas where deepwater fisheries are commercially viable, they comprise most of the exploited biomass. The demersal fishes are a diverse group and, for example, on the continental slope of the North Atlantic they are represented by some 57 families and at least 296 species (Merrett and Haedrich 1997). On the western Australian slope Williams, Koslow and Last (2001) recorded 108 families and 388 demersal species at depths between 200 and 1 460 m, but in other areas of the Pacific the numbers were considerably less (18-19 families and 85-199 species). As will become apparent below, relatively few of these species occur in sufficient quantities or are of a large enough body size to be of any commercial interest.

Benthic fishes were described above as those that are adapted to a life closely associated with the bottom. However, the distinction between benthic and benthopelagic is sometimes made on the basis of diet and can sometimes result in species of the same genus (e.g. Coryphaenoides, family Macrouridae) being classified in the different categories (Koslow 1996). Koslow also recognized that there was another group of demersal fishes that aggregate in association with seamounts and other rugged topography. Examples of such "seamount-associated" fishes are orange roughy, armourhead and alfosinos (Beryx spp.). Koslow (1996) compared the metabolism and life-history patterns of these aggregating species with those of the dispersed benthic and benthopelagic species. The seamount associated fishes had higher metabolic rates and flesh with high protein and lipid contents and a low water content. The high quality of the flesh results in a high market value compared with other deepwater species.

This paper is not intended to be a comprehensive review of deepwater fish and their environment. It is thirty years since I caught my first deep-sea fish in the Rockall Trough, North East Atlantic, and I make no apologies for drawing heavily on my long experience and interest in that area to find examples to answer many of the questions. It should be read in conjunction with Gordon et al. (1995) because I have placed most emphasis on recent advances in our knowledge of the fishes and their habitat.

2. HOW DO THE PHYSICAL FEATURES OF THE CONTINENTAL SLOPE, SEAMOUNTS AND OCEAN RIDGES COMPARE WITH THE SHELF?

Deepwater fisheries can occur on the continental slopes, around oceanic islands and on and around seamounts or ocean ridges. The characteristics of some of these areas are shown in Table 1.

TABLE 1
Comparison of some physical features of the shelf, continental slope, seamounts and ocean ridges.

Seamounts are underwater mountains with heights of more than 1 000 m. Those between 500 and 1 000 m and less than 500 m are known as knolls and hills respectively. It is estimated that there are about 30 000 seamounts in the Pacific Ocean, about 1 000 in the Atlantic Ocean and an unknown number in the Indian Ocean (Morato 2003).

In our 1995 paper (Gordon et al. 1995) we showed how the new technology of satellite navigation and track plotters had greatly increased the efficiency of the deepwater fishing process. Since then technology has continued to advance with greatly improved fish detection and trawl net monitoring systems, which in conjunction with high resolution mapping (swath bathymetry) means that there are fewer refuges for fish in the world’s oceans. The Fishing held in Vigo 2003 refers to "surgical" fishing technologies - i.e. a precision targeted trawling. From the conservation viewpoint a positive aspect could be the use of the same technology to avoid unwanted catch and reduce incidental impact on benthic environments. The same report noted that fishing on steep seamounts was not possible with current technology as 20-30^o slopes are the maximum that currently used gears can handle.

We also drew attention to the fact that increasing efficiency (often referred to as technology creep) has implications for catch and effort statistics. Many of the stock assessments of the deepwater species of the northeastern Atlantic (ICES area) that have been carried out since 1998, are based on French commercial trawl catch per unit of effort (CPUE) data. These data were selected from a sector of the French fleet that had used similar trawling techniques since the start of the fishery (Lorance and Dupouy 2001). Great care has to be taken when using time-series of CPUE data that make no allowance for increasing efficiency during the development of the fishery.

Research surveys often use different trawl gears to target specific species or to sample differing bottom topographies. A study of the deepwater fishing impacts in the Rockall Trough demonstrated the difficulty of combining data from different trawls (and depths) into a unified time series of CPUE (Basson et al. 2002). In fact it was only possible to compare pre- and post-fishery data for one species, the roundnose grenadier, and two broad groupings of exploited and unexploited species.

The technology of longlining has also continued to improve. For example, Bergstad and Hareide (1996) have documented a century of development in the Norwegian longline industry. Reis et al. (2001) has described changes in the black scabbardfish fishery off Madeira and shown how the increasing efficiency and effort has resulted in a decline in CPUE. The systems for automatically shooting and hauling lines are becoming more widely used and automatic baiting, and even the automated replacement of damaged hooks, is increasing the overall efficiency of the operation. The increasing accessibility of this technology is changing the patterns of many artisanal fisheries.

3. HOW DOES THE PHYSICAL ENVIRONMENT OF THE SLOPE, SEAMOUNTS AND OCEAN RIDGES DIFFER FROM THAT OF THE SHELF?

In our 1995 paper we described the physical environment of the continental slope with particular reference to the long time series of hydrographic studies in the Rockall Trough. At these temperate latitudes a seasonal thermocline forms during the summer and stratifies the water column. The breakdown of the thermocline during the winter and early spring redistributes the nutrients that fuel the seasonal blooms of phytoplankton production. Almost all the energy reaching the deep sea is derived from this surface primary production.

While the seasonal breakdown of the thermocline is an important factor at temperate latitudes it is not the only factor responsible for enhanced primary productivity along continental margins. In the tropics and sub-tropics the water column is usually permanently stratified and surface productivity is low. However, in areas where there are strong offshore winds surface water is displaced offshore and is replaced by colder, nutrient rich water from below the thermocline, which results in high, often seasonal, primary productivity. This phenomenon is known as upwelling and there are five major areas of coastal upwelling; off Peru and Chile, off California, off North West Africa, off Namibia and South West Africa and in the northwestern Arabian Sea. Apart from these major areas, localised upwelling can arise at the fronts separating major water masses or in association with eddies.

The waters above and around seamounts can be highly productive and support (or have supported) major fisheries for aggregating species such as armourhead, orange roughy and alfonsino. Oceanic currents impinging on the seamount and causing localised upwelling ("Taylor" columns) have often been associated with the high productivity of seamounts (Koslow 2001). However, it now appears that this alone would be inadequate to sustain the observed fish aggregations. It is more likely that it is the diurnal vertical migration of organisms being intercepted by seamounts and their physical aggregation by currents that provides the food to sustain most of the fisheries (Koslow 1997, 2001; Rogers 1994).

There is often enhanced biological enrichment around oceanic islands caused mainly by the interruption of the current flow forming eddies and a wake of disturbed flow (Barton 2001). Around higher islands the disruption of wind flow may also result in upwelling of deepwater.

Most established or developing deepwater fisheries are in areas where there is high surface productivity caused by one of the mechanisms described above. However, the existence of an exploitable biomass ultimately depends on this energy reaching the deep sea. There are some notable exceptions to this generalization. This topic is discussed in more detail in Section 6.

4. HOW DO THE DEMERSAL SPECIES ASSEMBLAGES ON THE SHELF AND SLOPE DIFFER FROM ONE ANOTHER AND ARE THE LATTER BASICALLY DIFFERENT FROM THE PELAGIC OCEANIC ASSEMBLAGE?

In our answer to this question in 1995, we divided the oceanic pelagic fauna into two groups, above and below 400 m. The shallower pelagic fauna had close affinities with shelf fishes, while the deeper pelagic species had a quite separate composition and might be termed true deep-sea species. Comparing the two pelagic zones with the demersal showed that there was little overlap of fish species. Only one pelagic family, the myctophids, has been commercially exploited. The commercially exploited species of the demersal assemblage comprise only a small part of the total species richness. The important families include the Squalidae (sharks), Rajidae (skates), Macrouridae (grenadiers), Gadidae (lings and forkbeards), Moridae (morids), Lophiidae (anglerfish), Trachithyidae (orange roughy), Scorpaenidae (redfish and scorpionfishes) and Pleuronectidae (Greenland halibut)

The high diversity of the slope fishes has important implications for fisheries where only relatively few species are of commercial value. Fishing gear selectivity is important if high discard rates are to be avoided. Few if any fish brought to the surface and subsequently discarded will survive. It is also generally considered that there will be a high mortality of fish escaping through the meshes of trawl nets while being towed on the bottom because they have fragile skins lacking in mucus. It is doubtful if the increasing use of trawl selectivity methods such as square mesh panels or use of grids in shallow waters will, if used in deepwater, have any significant conservation value if the fish that escape do not survive. Longlining is often promoted as a more selective fishing method but key commercial species, such as orange roughy and roundnose grenadier, are not caught by this method and the high bycatch of sharks is a major concern.

In well studied shallow-water areas, such as the North Sea or the Gulf of Maine, we have a good overall impression of the inter- and intra-annual changes in the structure of the total fish assemblage and how it varies with habitat and by depth. This knowledge is greatly enhanced by the use of many different fishing gears all of which catch a different portion of the total assemblage. In the deep sea the amount of sampling has been relatively small and has often been limited by season, depth or gear type. Even in an area such as the continental margin to the west of the British Isles (Rockall Trough) that has been scientifically sampled for more than 100 years (Gordon 2003), our concept of the assemblage may be biased by the use of relatively few types of sampling gear that have been used. The most frequently used sampling method has been the bottom trawl (beam or otter) and this immediately puts a constraint on the type of habitat that can be sampled, i.e. a fairly flat and relatively smooth seabed.

In the Rockall Trough we have used three different research trawls to sample the fish assemblages. One was similar to a commercial trawl, but with a smaller overall mesh size; the other was a small shrimp trawl which was fished either on single or paired warps. Table 2 shows the five top ranking species in the 1 000 m bathymetric zone expressed as a percentage of abundance and of biomass. This shows quite clearly the differences in catchability between gears. It is also apparent that relatively few species comprise a high proportion of the total abundance and biomass. This also applies to all other bathymetric zones surveyed in the Rockall Trough (Gordon and Bergstad 1992) and in the Porcupine Seabight (SW Ireland) (Gordon et al. 1996).

There have been relatively few longline surveys in the Rockall Trough where the total catch has been recorded (Connolly and Kelly 1996, Stene and Buner 1991). In contrast to the bottom trawls the catches were almost entirely composed of gadoids, morids and deepwater sharks.

TABLE 2
The relative numerical abundance and biomass of the top five species in the 1 000 m bathymetric zone of the Rockall Trough by gear type

5. HOW DO THE BASIC DISTRIBUTION PATTERNS (VERTICAL AND HORIZONTAL) OF DEEPWATER FISHES COMPARE WITH THEIR SHELF-DWELLING COUNTERPARTS?

In the 1995 paper we discussed the vertical and horizontal distribution patterns in the North Atlantic based on the atlas of deepwater demersal fishes of the North Atlantic compiled by Haedrich and Merrett (1988). We noted that there were considerable differences in the depth range of individual species, ranging from more than a thousand metres to a very restricted few hundreds of metres. Examples are the cutthroat eel (Synaphobranchus kaupi), which has a very large depth range and also exhibits a well marked ‘bigger-deeper’ trend (Gordon and Mauchline 1996) and the tripod fish (Bathypterois dubius) which has a very restricted range (Merrett et al. 1991a). The ‘bigger-deeper’ phenomenon, where juveniles tend to live at shallower depths than adults is of widespread occurrence in deepwater demersal fishes (e.g. Merrett et al. 1991b). There was no evidence to support any pronounced zonation of deepwater demersal fishes. Instead, there was a gradual replacement of species although in a number of investigations the rate of change tended to be greatest at around 2000 m depth (Merrett and Haedrich 1997). However, zonation of fish assemblages can occur and is usually associated with physical phenomena. For example, on the continental slope off Norway there is a relatively sudden transition from warmer Atlantic water to cold Arctic water with virtually no similarity between the fish faunas of the two water masses (Bergstad et al. 1999).

Gordon et al. (1995) noted the importance of a knowledge of depth distributions for the rational exploitation of deepwater species. Commercial fishing may not affect fish in the whole depth distribution of a species and this needs to be taken into account when carrying out stock assessments. The overlapping depth ranges of many species makes it difficult to select for target species and avoid an unwanted bycatch on non-target species. Discard levels are generally high in the mixed fisheries of the northeastern Atlantic and can amount to up to 50 percent of the total catch by weight at some depths. Data from research trawls with small mesh can be used to estimate the quantity of escapees from commercial trawls. In the Rockall Trough these have been estimated to be from about 66 to 86 percent by number and 10 to 45 percent by weight of fish entering the trawl (Gordon 2003). It is probable that there will be a high mortality among these escapees.

The number of species in any given depth zone also changes with depth and usually decreases rapidly below about 1500 m. Gordon and Mauchline (1990) estimated total fish abundance and biomass by combining the results from different gears used in the Rockall Trough. They found a clear peak in both abundance and biomass of demersal fishes at mid-slope depths (1 000-1 500 m). This phenomenon has been observed in many other areas and the probable link to food chains will be discussed in Section 6. These peaks are not universal and, for example, in the Norwegian basin demersal fish biomass decreases exponentially with depth (Bergstad et al. 1999). Off Western Australia there was little change in demersal fish biomass with depth (Williams, Koslow and Last 2001).

We also noted that the vertical distribution of deepwater demersal species in the water column was poorly understood. This was a result of the difficulty of using large midwater trawls at depth. There has been little progress in this field and it remains a matter of conjecture whether some of the unknown life history stages of species such as black scabbardfish and deepwater sharks might occur in midwater.

One area where there have been significant advances since 1994 is in the field of in situ observation of deep-sea fishes. Priede and Bagley (2000) have reviewed the use of autonomous unmanned landers for in situ studies of behaviour. While much of the deep-sea pioneering work was carried out at abyssal depths there have recently been investigations of continental slopes, for example in the Porcupine Seabight (northeastern Atlantic) (Priede et al. 1994, Bagley et al. 1994), on the Patagonian slope (Collins et al. 1999) and in the Mediterranean (Jones 1999)

While the investigations of fish behaviour using landers mainly relate to species that are attracted to bait, observations from manned submersibles or remotely operated vehicles (ROVs) provide valuable insights into the relationship between the fish and their habitat. However, fish disturbance by noise or light while being observed can be a significant problem.

In 1996 and 1998 the French manned submersibles Cyana and Nautile were deployed on the continental slope of the Bay of Biscay at depths between 400 and 2 000 m. Their observations have resulted in new information on fine-scale habitat selection and behaviour of several deepwater species (Lorance, Latrouite and Sécet 2000, Lorance et al. 2002, Uiblein et al. 2002, 2003)

In August 2002 the French ROV (Victor 6000) was used to visually estimate demersal fish abundance at three contrasting areas at depths between 1 200 and 1 500 m in the Bay of Biscay. The areas differed in topography, current conditions and previous fishing activity. The abundance estimates were compared with those estimated from a baited camera and from the catch of a commercial trawler that fished the same area after the ROV transect (Trenkel et al. 2002). The visual observations provided a wealth of information on fish behaviour in relation to habitat.

6. HOW DOES THE VERTICAL DISTRIBUTION PATTERN OF FISH BIOMASS CORRELATE WITH THE TROPHIC INPUT TO THE OCEANIC ENVIRONMENT?

The source of food for deepwater fishes is almost entirely derived from primary production in the euphotic zone. An exception is the specialized fauna associated with chemosynthesis around hydrothermal vents. The food chain from phytoplankton, herbivores, carnivores and ultimately to deep-sea demersal fishes can vary in complexity and as the number of stages in the food chain increases so the energy available to demersal fish decreases. The classical concept of a rain of detritus and, or, overlapping pelagic food chains implies in a decrease in the biomass of plankton, micronekton and benthos with increasing depth.

However, as noted above many studies have shown that the biomass of demersal fish often peaks at midslope depths of around 1 000 to 1 500 m (Gordon and Mauchline 1990, Merrett et al. 1991a, Koslow et al. 1994). Unless the turnover rates of benthos are high, and there is no evidence that they are, then the benthic biomass on the midslope could not support the observed biomass of demersal fishes (Gordon et al. 1995)

Gordon et al. (1995) listed the following factors that might contribute to the enhanced demersal fish biomass; an increase in primary production along the shelf-slope break, slope currents and tidal effects and impingement of pelagic organisms, both horizontal and vertical, around the oceanic rim. There have been many studies on the diets of deepwater fish species (see Gartner et al. 1997) for a review) and these show that in most areas where there are exploited, or potentially exploitable, demersal fish, their diet consists predominantly of pelagic or benthopelagic fish and invertebrates. There is increasing evidence that it is the impingement, vertical or horizontal, of the vertically migrating mesopelagic fauna onto the slope (e.g Mauchline and Gordon 1991, Williams and Koslow 1997) or around seamounts (e.g Rogers 1994, Koslow 1997) that sustains the high densities of exploitable fishes on the midslope. The daily vertical migration of midwater organisms therefore provides the energy required to sustain deepwater fisheries.

The important fisheries of the shallow continental shelves are all in areas where there is high surface productivity such as mid-latitudes, upwelling areas or around oceanic islands. This high productivity is fuelled by the essential nutrients being brought to the euphotic zone by winter mixing of the water column, upwelling of deep, nutrient rich water, or the interruption of ocean currents around islands. If the important deepwater fisheries depend on the efficient transfer of energy produced at the surface then these fisheries should also be in areas where there is high surface productivity.

Perhaps the best known example of the link between surface productivity and deepwater fish communities relates to abyssal fish in the North Atlantic. The fish communities at about 4 000 to 5 000 m depth on the Porcupine and the Madeira Abyssal Plains were compared by Merrett (1987, 1992). In the Porcupine area there is a seasonal thermocline and its breakdown in the winter provides nutrients for the spring phytoplankton bloom. There is a permanent thermocline over the Madeira Abyssal Plain and the low nutrient levels result in low surface production. The effect of these differing features has quite a dramatic effect on the demersal fish communities. The fish of the Madeira Abyssal Plain are of small adult body size, negatively buoyant and feed mainly on epibenthic or benthic organisms. By contrast the fish of the Porcupine Abyssal Plain are of large body size, neutrally buoyant with greater mobility and are predominantly benthopelagic feeders. It is the latter type of benthopelagic fish that is exploited on the slopes but the biomass at these abyssal depths is low and would not support a fishery.

If we now extrapolate these findings to the continental slopes we find that most of the existing deepwater fisheries occur in areas of high surface productivity. Most of these fisheries are on the upper and mid-slopes down to about 1 000 to 1 500 m where there is a peak of demersal fish biomass (see Section 5). This corresponds to the daytime depth of the vertically migrating mesopelagic fauna supporting the hypothesis that it is the efficient transfer of energy from productive surface waters by the impingement of the mesopelagic fauna on the slope of the continental margin, seamounts or islands that sustains the fisheries.

In the North Atlantic the most productive deepwater fisheries are along the highly productive oceanic rim from about the Gulf of Maine around to the Iberian Peninsula. The seamounts of the Reykjanes Ridge and the Mid-Atlantic Ridge have been exploited for many years and there are deepwater fisheries around the oceanic islands such as Madeira and the Azores. There have been several research surveys of the continental slope off the eastern United States (Haedrich and Merrett 1988, references cited therein). There is decreasing productivity from north to south and hence no significant deepwater resources. Trawl and submersible survey investigations on the slope off Cape Hatteras (35 ºN) revealed a strikingly different fish fauna from that of a nearby area, comprising small-sized individuals, which Sulak and Ross (1996) termed ‘Lilliputian’. There was a high fish density and a low number of benthopelagic feeders reminiscent of an area with low surface productivity. However, the surface productivity was similar between the two areas and the suggested explanation for the Lilliputian fauna was that the Hatteras middle slope has an unusual hydrographic convergence resulting in virtually no net current flow. The organic flux from surface waters accumulates in the area and results in episodic hypoxia at the sediment surface.

The high productivity of the Norwegian Sea supports important shelf and oceanic epipelagic fisheries, but the deeper continental slope, comprising cold Norwegian Deepwater, has no important fisheries. At the transition between the warmer Atlantic water and the cold water there are fisheries for some species, notably Greenland halibut and redfish (Sebastes species) (Bullough et al. 1998).

There are no significant deepwater fisheries in the sub-tropical or tropical North Atlantic. There is good descriptive information on the fish assemblages of the West African slope including the upwelling area (Golovan 1978, Merrett and Marshall 1981, Merrett and Domanski 1985) but detailed quantitative data is lacking. It is possible that the initial Russian exploratory fishing never developed into a fishery because of the dominance of alepocephalid fishes in the catch (Golovan and Pakhorukov 1975). These fishes have a watery flesh and are of little interest for human consumption.

The Mediterranean is a subtropical semi-enclosed sea and is unusual in having a very stable temperature of about 13 ºC from below the thermocline to abyssal depths. Surface productivity is generally low but is elevated in frontal areas such as the Balearic Basin. Deepwater crustacean fisheries for high value species, such as Aristeus antennatus and Aristaeomorpha folliacea, are important. The fish bycatch is generally quite low. The main deepwater fisheries are on the Catalan Slope in the Balearic Basin, and the Ionian Sea.

In the South Atlantic the deepwater fisheries are centred on the slope off South West Africa and on the edge of the Patagonian shelf. Off South West Africa, in the Benguella upwelling, the slope fisheries are dominated by the Cape hakes, kingclip (Genypterus capenensis) and in recent years orange roughy. The high seasonal productivity of the Patagonian shelf extends onto the slope where the dominant catches are of southern blue whiting (Micromesistius australis), the Patagonian Grenadier^[30] (Macruronus magellanicus), the pink cusk eel (Genypterus blacodes) and, more recently, the Patagonian toothfish (Dissostichus eleginoides). The potential of deepwater fishing off Southern Brazil in the area of the sub-tropical convergence has recently been investigated (Perez et al. 2003).

Although there is high seasonal surface productivity in the northern part of the North West Pacific the only important deepwater fisheries are on the upper slope for Pacific ocean perch (Sebastes alutus) and some scorpion fishes. Recent surveys of the deeper slope off the northern Kuril Islands, and southeastern Kamchatka suggest a fauna reminiscent of the Norwegian Sea dominated by benthic feeding fishes such as lycodids (Orlov 2003, Tokranov and Orlov 2002). The slope of the northeast Pacific, which is influenced by the high seasonal productivity in the north and the Californian upwelling in the south, has more important deepwater fisheries dominated by the scorpaenid fishes of the genus Sebastes and Sebastolobus, the sablefish and the Dover sole (Microstomus pacificus). Deepwater fisheries on concentrations of the armourhead and the alfonsino (Beryx splendens) around seamounts in the Central Pacific were heavily exploited in the 1970s to the extent that they are now virtually commercially extinct. In the southwest Pacific the productive waters, characterized by Antarctic Intermediate Water, around New Zealand and its extensive underwater plateaus and also off southeastern Australia support important deepwater fisheries for southern blue whiting, blue grenadier (Macruronus novaezelandiae), oreosomatids and orange roughy. The Chilean upwelling area of the southeast Pacific also supports considerable deepwater fisheries for Patagonian grenadier, southern blue whiting and Patagonian toothfish.

The only deepwater demersal fishery of any significance in the Indian Ocean was the short-lived fishery for orange roughy that developed around seamounts in international waters (FAO 2001, 2002). A scientific study of the fish communities off western Australia (Williams, Koslow and Last 2001) is of interest in relation to the link between surface productivity and the potential for slope fisheries. The low overlying productivity caused by downwelling of water, initially low in nutrients, results in a diverse fauna of small, typically benthic species and with no evidence of the midslope peak of biomass characteristic of exploited areas.

In the Southern Ocean the important demersal deepwater species is the Patagonian toothfish. Although the overall seasonal surface productivity is high in this area the fishery tends to be concentrated around islands such as the Falklands, South Georgia, Heard, Macdonald and Kerguelen. However some areas with high production such as the Antarctic peninsula are unproductive for fishers, especially toothfish (D. Agnew, MRAG, London; pers. comm.).

7. WHAT IS KNOWN ABOUT DEEP-SEA FISH POPULATION STRUCTURE AND BREEDING BIOLOGY?

Gordon et al. (1995) noted that bimodal length distributions were a common feature of deepwater fish and invertebrate populations. This arises in long-lived and/or unexploited populations where the exponential decrease in growth rate with increasing age results in a stacking of older age classes by size class.

The age of fish is generally estimated by counting growth rings on hard parts such as scales, otoliths, vertebrae or fin rays. In shallow-water temperate latitudes these growth zones tend to be annual and can be validated, by tagging, chemical marking, daily growth rings, edge analysis, etc. The basis for the zones is the differential seasonal growth rates resulting from the direct effect on metabolism of changes in physical variables such as temperature or day length or indirectly by the effect of these parameters on food supply and, or, quality. Growth zones or checks can also result from the influence of other factors that affect the metabolism of the fish, such as spawning. For the true deepwater fish that spend their whole life cycle in an environment of constant darkness and temperature it is interesting that most have well marked growth rings. The reason for changes in the growth of the hard parts of these deepwater fishes is unknown but is probably related to either the availability, or the quality, of their food. Validation of the annual nature of the growth zones in deepwater fish is difficult because live fish are not available for tagging or marking experiments. Age estimation in orange roughy, generally considered to be a slow-growing, long-lived species, has been reviewed by Tracey and Horn (1999). The ages of juvenile fish, up to three years, have been validated by comparing seasonal changes in the type of growth at the otolith margin and peaks in the length frequency distributions of juvenile fish. Otolith margin analysis has also been used to validate the annual nature of growth zones of juveniles of several macrourid fishes (Coggan, Gordon and Merrett 1999, Gordon and Swan 1996, Morales-Nin 2001, Swan and Gordon 2001).

The extrapolation of the growth zones of juvenile fish to adult fish can be difficult as the zones become narrower and tightly packed and the otolith shape changes with growth. To view these growth zones it is often necessary to section the otolith. Changes in the growth axis with age can make the growth zones of the otolith difficult to interpret, as was shown for larger-sized roundnose grenadier by Bergstad (1990). Where the growth zones can be counted in whole otoliths there can be a discrepancy between these counts and those from sections. A good illustration of this is in the black scabbardfish where higher ages are estimated from sections (Morales-Nin et al. 2002). However, in this study the interpretation was made more difficult because of a lack of juvenile specimens, no clear seasonal growth patterns at the otolith margins and the possibility of spatial/stock differences. It was considered that sectioning may have revealed additional, non-annual growth zones leading to an overestimation of age.

The great longevity of orange roughy (up to about 140 years) is controversial. Smith et al. (1995) demonstrated the differences in age estimates between whole and sectioned otoliths and using radiometric techniques obtained ages comparable with those obtained from sectioned otoliths. Radiometric ageing, which uses the disequilibria between 210Pb and 226Ra, had previously been used to estimate the ages of long-lived, shallow-water species (See Gordon 1998 for a review). However, the radioactive decay process involves the gas Rn and a key assumption has been that that the otoliths have been impervious to gas loss. Gauldie and Cremer (2000) have demonstrated gas loss from orange roughy otoliths and questioned the validity of the high age estimates. Radiometric ageing has also been used by Andrews et al. (1999) to validate ageing of the Pacific grenadier (Coryphaenoides acrolepis). They addressed the problem of Rn loss and suggested that it did not occur in vivo. Kastelle and Forsberg (2002) found evidence, although not conclusive, that Rn did not escape from Pacific halibut otoliths.

Although the "jury is still out" on the question of particular old ages of deepwater species it is worth noting that the old ages often cited in the literature are the maximum reported ages. For example, although ages of up to 70 years have been reported for roundnose grenadier most of the fish in the commercial landings are less than 30 years.

In our 1995 paper we noted that fishing down the accumulated biomass of larger fish with a high age and/or size at first maturity could have important consequences in reducing recruitment. We also noted that there was evidence to suggest that some deepwater fishes may not breed every year. Many shelf species attain maturity while they are growing relatively fast and we suggested that deepwater fishes might channel their resources into reproduction only after growth had effectively ceased. Merrett (1994) comprehensively reviewed the breeding biology of deepwater species of the North Atlantic and although there have been new studies on individual species there is still a lack of basic information. The eggs and larvae of deepwater fishes and their distribution in the water column continues to be poorly understood.

8. CONCLUSIONS

8.1 Validity of generalizations

Deepwater fishes are long lived and slow growing, have a high age and large size at first maturity and have a low fecundity. This sentence, or a variant of it, is frequently used in the context of deepwater fisheries. Almost invariably, the statement is coupled with the orange roughy. The question that is seldom addressed is how valid is this generalization? In the context of the mixed trawl fisheries of the northeast Atlantic, ICES asked its Working Group on the Biology and Assessment of Deep-sea Fishery Resources to rank the main species in terms of vulnerability relative to two species with a longer history of exploitation, redfish and Greenland halibut. The criteria for this exercise were longevity, growth, natural mortality, fecundity and length or age at first maturity where data were available. Table 3 summarizes the conclusions (the full details for each of the criteria and the sources are published in ICES (2001)). ICES emphasizes that the underlying data are of variable quality but, nevertheless, believe that the main pattern is robust. These data indicate, that at least for the mixed fisheries, we should be cautious about making broad generalizations about the life history patterns of deepwater fishes.

TABLE 3
Ranked vulnerability of deepwater species based on life history parameters using redfish and Greenland halibut as reference species. 1 - most vulnerable, 5 - least vulnerable (Modified from ICES (2001).

Are there any additional questions that have arisen in the intervening years that need to be addressed? There are probably two that warrant increased research effort, which I address in the following sections.

8.2 Stock identity

The first question is related to management/assessment units and stock identity. Often, the statistical data on landings and effort are provided for areas that have little relevance to the distribution of the stock. This has been a particular problem in the northeast Atlantic where the long established ICES statistical Sub-areas and Divisions were devised for shelf fisheries and are inappropriate for deepwater species. Assessments are carried out on statistical units that in many cases have little relevance to the biological stock. With the notable exception of orange roughy in the South Pacific (Smith et al. 2002) there have been surprisingly few studies on the genetics of deepwater species. Many deepwater species have wide, even global, distributions and for effective management more information is required on stock structure so as to define appropriate management units. New technologies such as otolith microchemistry might provide additional information on stock discrimination (Edmonds et al. 1991, Swan et al. 2003a,b)

8.3 Environmental effects of fishing

There are increasing concerns about the ecosystem effects of fishing activity, made all the more obvious by the new technologies described in Section 5. Koslow et al. (2000) reviewed the available information on the impacts of deepwater fishing and drew attention to the lack of knowledge of the effects at the level of fish assemblages and on predator-prey relations. The recent study of the pre- and post-fishery impacts of deepwater fishing on the fish communities in the Rockall Trough has yielded new insights and also highlighted some of the problems associated with the use of historical data (Basson et al. 2002). Gordon (2003) summarized the available information on fishing impacts in the Rockall Trough. The global perspective is given by Koslow et al. (2000), who drew attention to the impacts on deepwater habitats, especially seamounts. The destruction of cold-water corals and their associated fauna by fishing gears has become a sensitive issue so that many countries, including New Zealand, Australia, Canada, Norway and most recently the European Union, have acted to protect some areas. However, there has been little research on the effect of trawling on deepwater soft sediments (Cryer et al. 2002) yet we know that the visible effects of bottom trawling are all too evident (Roberts et al. 2000, Trenkel et al. 2002). The effects of the removal of top predators and the frequently high levels of discarding on the ecosystem and for biodiversity are largely unknown.

8.4 The need for caution

The questions posed in 1994 remain equally as valid today and in revisiting them it is clear that there have been some considerable advances in our knowledge of the biology of the individual fish, the assemblages and their role in the ecosystem. It is also evident that the deepwater fisheries have developed considerably in recent years and, in many instances, there are serious concerns about their future sustainability. Haedrich, Merrett and O’Dea 2001) demonstrated how science has lagged behind the boom and bust of the fishery and unfortunately this is still all too often the case today. Food supply, which decreases with depth, influences most of the life history traits of the deepwater species. Deepwater fish populations cannot be expected to support the levels of exploitation that have been applied to shelf populations.

9. ACKNOWLEDGEMENTS

I thank the Scottish Association for Marine Science for their support of my programme of deepwater fish research over many years. I am also grateful for their continuing support as a retired honorary fellow. I also acknowledge the support of the European Commission for part funding many research projects and in particular EC FAIR 95/655 Developing deepwater fisheries: data for their assessment and for understanding their interaction with and impact on a fragile environment, which gave me so much satisfaction, enjoyment and long-lasting friendships. There are too many people who have helped along the road to give individual thanks, but I would single out John Mauchline who gave me so much early encouragement and in later years was a pleasure to work with. Janet Duncan and Sarah Swan gave me valuable and good humoured support for 19 and 10 years respectively and to them I owe a debt of gratitude. Finally, underpinning all my research was RRS Challenger, all her crew and the numerous participating scientists. I had the privilege of being a ‘Rockall Ranger’ on her first and last fishing trips (1973 and 1999) and experiencing, in the early days, the ‘simple life’ of seagoing before the advent of satellite phones and e-mails made it similar to another day in the office.

10. LITERATURE CITED

Andrews, A.H., G.M. Cailliet & K.H. Coale 1999. Age and growth of the Pacific grenadier (Coryphaenoides acrolepis) with age estimates validation using an improved radiometric ageing technique. Canadian Journal of Fisheries and Aquatic Science. 56: 1339-1350.

Bagley, P.M., A. Smith & I.G. Priede 1994. Tracking movements of deep demersal fishes in the Porcupine Seabight, north-east Atlantic Ocean. Journal of the Marine Biological Association of the United Kingdom. 74: 473-480.

Barton, E.D. 2001. Island wakes. In: J.H. Steele and S.A. Thorpe (Eds) Encyclopedia of Ocean Sciences. Academic Press. pp. 1397-1403.

Basson, M., J.D.M. Gordon, P. Large, P. Lorance, J. Pope & B. Rackham 2002. The effects of fishing on deepwater fish species to the west of Britain. JNCC Report. No. 324: 150 pp.

Bergstad, O.A. 1990. Distribution, population structure, growth and reproduction of the roundnose grenadier Coryphaenoides rupestris (Pisces: Macrouridae) in the deepwaters of the Skagerrak. Marine Biology. 107: 25-39.

Bergstad, O.A. & N.R. Hareide 1996. Ling, blue ling and tusk of the north-east Atlantic. Fisken og havet. no.15: 126p.

Bergstad, O.A., O. Bjelland & J.D.M. Gordon 1999. Fish communities on the slope of the eastern Norwegian Sea. Sarsia. 84: 67-78.

Bullough, L.W., W.R. Turrell, P. Buchan & I.G. Priede 1998. Commercial deepwater trawling at sub-zero temperatures - observations from the Faroe-Shetland Channel. Fisheries Research. 39: 33-41.

Clark, M. 2001. Are deepwater fisheries sustainable? - the example of orange roughy (Hoplostethus atlanticus) in New Zealand. Fisheries Research. 51: 123-135.

Coggan, R.A., J.D.M. Gordon & N.R. Merrett 1999. Aspects of the biology of Nezumia aequalis from the continental slope west of the British Isles. Journal of Fish Biology. 54: 152-170.

Collins, M.A., I.G. Priede & P.M. Bagley 1999. In situ comparison of activity in two deep-sea scavenging fishes occupying different depth zones. Proceedings of the Royal Society of London (B). 266: 2011-2016.

Connolly, P.L. & C.J. Kelly 1996. Catch and discards from experimental trawl and longline fishing in the deepwater of the Rockall Trough. Journal of Fish Biology. 49 (Supplement A): 132-144.

Cryer, M., B. Harthill & S. O’Shea 2002. Modification of marine benthos by trawling: Toward a generalization for the deep ocean? Ecological Applications. 12: 1824-1839.

Edmonds, J.S., N. Caputi & M. Morita 1991. Stock discrimination by trace-element analysis of otoliths of orange roughy (Hoplostethus atlanticus), a deepwater marine teleost. Australian Journal of Marine and Freshwater Research. 42: 383-389.

FAO 2001. Report of the Ad Hoc Meeting on Management of Deepwater Fisheries Resources of the Southern Indian Ocean, Swakopmund, Namibia, 30 May-1 June 2001. FAO Fish. Rep. 652. FIRM, Rome. 61pp.

FAO 2002. Report of the Second Ad Hoc Meeting on Management of Deepwater Fisheries Resources of the Southern Indian Ocean, Fremantle, Western Australa, 20 - 22 May 2002. FAO Fish. Rep. 677. FIRM, Rome. 106pp.

Garibaldi, L. & L. Limongelli 2003. Trends in oceanic captures and clustering of large marine ecosystems. Two studies based on the FAO capture database. FAO Fisheries Technical Paper. 435: 71 pp.

Gartner, J.V., Jr., R.E. Crabtree & K.J. Sulak 1997. Feeding at depth. In: D.J. Randall and A.P. Farrell (Eds) Deep-Sea Fishes. Academic Press, San Diego. pp. 115-194.

Gauldie, R.W. & M.D. Cremer 2000. Confirmation of 222Rn loss from otoliths of orange roughy Hoplostethus atlanticus. Fisheries Research. 66: 989-991.

Gjøsaeter, J. & K. Kawaguchi 1980. A review of the world resources of mesopelagic fish. FAO Fisheries Technical Paper. 193: 151pp.

Golovan, G.A. 1978. Composition and distribution of the ichthyofauna of the continental slope of North-Western Africa. Trudy Instituta Okeanologiya. 111: 195-256. (In Russian)

Golovan, G.A. & N.P. Pakhorukov 1975. Some data on the morphology and ecology of Alepocephalus bairdii (Alepocephalidae) of the central and eastern Atlantic. Journal of Ichthyology. 15: 44-50.

Gordon, J.D.M. 1998. Radiometric ageing. In: P.J. Wright (Ed.) The present status of otolith research and applications. European Commission EFAN Report 1-98. pp. 22-31.

Gordon, J.D.M. 2001a. Open ocean fisheries for deepwater species. In: J.H. Steele, S.A. Thorpe and K.K. Turkekian (Eds) Encyclopedia of Ocean Sciences. Academic Press. pp. 2023-2030.

Gordon, J.D.M. (Ed.) 2001b. Deepwater fish and fisheries. Fisheries Research. 51. pp. 105- 412.

Gordon, J.D.M. 2003. The Rockall Trough, North East Atlantic: the cradle of deep-sea biological oceanography that is now being subjected to unsustainable fishing activity. Journal of Northwest Atlantic Fishery Science. 31: 57-83.

Gordon, J.D.M. & O.A. Bergstad 1992. Species composition of demersal fish in the Rockall Trough North-eastern Atlantic as determined by different trawls. Journal of the Marine Biological Association of the United Kingdom. 72: 213-230.

Gordon, J.D.M. & J. Mauchline 1990. Depth-related trends in diet of a deep-sea bottom-living fish assemblage of the Rockall Trough. In: M. Barnes and R.N. Gibson (Eds) Trophic Relationships in the Marine Environment. Proceedings of the 24^th European Marine Biology Symposium, Aberdeen University Press, Aberdeen. pp. 439-452.

Gordon, J.D.M. & J. Mauchline 1996. The distribution and diet of the dominant, slope dwelling eel, Synaphobranchus kaupi, of the Rockall Trough. Journal of the Marine Biological Association of the United Kingdom. 76: 493-503.

Gordon, J.D.M., N.R. Merrett, O.A. Bergstad & S.C. Swan 1996. A comparison of the deepwater demersal fish assemblages of the Rockall Trough and Porcupine Seabight, eastern North Atlantic: continental slope to rise. Journal of Fish Biology. 49 Supplement A: 217-238.

Gordon, J.D.M., N.R. Merrett & R.L. Haedrich 1995. Environmental and biological aspects of slope-dwelling fishes of the North Atlantic. In: A.G. Hopper (Ed.) Deepwater Fisheries of the North Atlantic Oceanic Slope. Kluwer Academic Publishers, The Netherlands. pp. 1-26.

Gordon, J.D.M. & S.C. Swan 1996. Validation of age readings from otoliths of juvenile roundnose grenadier, Coryphaenoides rupestris, a deepwater macrourid fish. Journal of Fish Biology. 49: 289-297.

Haedrich, R.L. & N.R. Merrett 1988. Summary atlas of deep-living demersal fishes in the North Atlantic Basin. Journal of Natural History. 22: 1325-1362.

Haedrich, R.L., N.R. Merrett & N.R. O‘Dea 2001. Can ecological knowledge catch up with deepwater fishing? a North Atlantic perspective. Fisheries Research. 51: 113-122.

Hopper, A.G. (Ed.) 1995. Deepwater fisheries of the North Atlantic oceanic slope. Kluwer Academic Press, Netherlands. pp. 420.

Humphreys, R. & R. Moffitt 1999. Unit 17 Western Pacific bottomfish and armorhead fisheries Our living oceans. Report on the status of U.S. living marine resources 1999. U.S. Department of Commerce, NOAA Technical Memo. pp. 4 pp.

ICES 2001. Report of the ICES Advisory Committee on Fishery Management 2001. ICES Cooperative Research Report. 246: 895 pp.

Jones, E.G. 1999. Burial at sea; consumption and dispersal of large fish and cetacean food-falls by deep-sea scavengers in the North East Atlantic Ocean and Eastern Mediterranean. The fate of cetacean carcasses in the deep sea. PhD-University of Aberdeen.

Kastelle, C.R. & J.E. Forsberg 2002. Testing for loss of Rn-222 from Pacific halibut (Hippoglossus stenolepis) otoliths. Fisheries Research. 57: 93-98.

Koslow, J.A. 1996. Energetic and life-history patterns of deep-sea benthic, benthopelagic and seamount-associated fish. Journal of Fish Biology. 49 Supplement A: 54-74.

Koslow, J.A. 1997. Seamounts and the ecology of deep-sea fisheries. American Scientist. 85: 168-176.

Koslow, J.A. 2001. Fish stocks and benthos of seamounts. In: H. Thiel and J.A. Koslow (Eds) Managing risks to biodiversity and the environment on the high sea, including tools such as marine protected areas - Scientific requirements and legal aspects. Proceedings of the Expert Workshop held at the International Academy for Nature Conservation Isle of Vilm, Germany, 27 February-4 March 2001. BfN-Skripten, 43: pp. 43-54.

Koslow, J.A., G.W. Boehlert, J.D.M. Gordon, R.L. Haedrich, P. Lorance & N. Parin 2000. Continental slope and deep-sea fisheries: implications for a fragile ecosystem. ICES Journal of Marine Science. 57: 548-557.

Koslow, J.A., C.M. Bulman & J.M. Lyle 1994. The mid-slope demersal fish community off southeastern Australia. Deep-Sea Research. 41: 113-141.

Lorance, P. and H. Dupouy 2001 CPUE abundance indices of the main target species of the French deepwater fishery in ICES Sub-areas V-VII. Fisheries Research. 51: 137-149.

Lorance, P., D. Latrouite & B. Séret 2000. Observations of chondrichthyan fishes (sharks, rays and chimaeras) in the Bay of Biscay (North-Eastern Atlantic) from submersibles. In: B. Séret and J.Y. Sire (Eds) Proceedings of 3rd European Elasmobranch Association Meeting, Boulogne-sur-Mer, 1999. Paris: Soc. Fr. Icthyol. + IRD 2000. pp. 29-45.

Lorance, P., F. Uiblein & D. Latrouite 2002. Habitat, behaviour and colour patterns of orange roughy Hoplostethus atlanticus (Pisces: Trachichthyidae) in the Bay of Biscay. Journal of the Marine Biological Association of the United Kingdom. 82: 321-331.

Mauchline, J. & J.D.M. Gordon 1991. Oceanic pelagic prey of benthopelagic fish in the benthic boundary layer of a marginal oceanic region. Marine Ecology Progress Series. 74: 109-115.

McIntyre, A.D. & J.E. Thorpe (Eds) 1996. Deepwater fishes Journal of Fish Biology, 49 Supplement A. pp. 1-348.

Merrett, N.R. 1987. A zone of faunal change in assemblages of abyssal demersal fish in the eastern North Atlantic; a response to seasonality in production? Biological Oceanography. 5: 137-151.

Merrett, N.R. 1992. Demersal ichthyofaunal distribution in the abyssal eastern North Atlantic, with special reference to Coryphaenoides (Nematonurus armatus) (Macrouridae). Journal of the Marine Biological Association of the United Kingdom. 72: 5-24.

Merrett, N.R. 1994. Reproduction in the North Atlantic oceanic ichthyofauna and the relationship between fecundity and species’ 207-245.

Merrett, N.R. & P.A. Domanski 1985. Observations on the ecology of deep-sea bottom-living fishes collected off Northwest Africa: 2. The Moroccan slope (27-34 ^oN), with special reference to Synaphobranchus kaupi. Biological Oceanography. 3: 349-399.

Merrett, N.R., J.D.M. Gordon, M. Stehmann & R.L. Haedrich 1991a. Deep demersal fish assemblage structure in the Porcupine Seabight (eastern North Atlantic): Slope sampling by three different trawls compared. Journal of the Marine Biological Association of the United Kingdom. 71: 329-358.

Merrett, N.R. & R.L. Haedrich 1997. Deep-sea demersal fish and fisheries. Chapman and Hall, London, 282 pp.

Merrett, N.R., R.L. Haedrich, J.D.M. Gordon & M. Stehmann 1991b. Deep demersal fish assemblage structure in the Porcupine Seabight (eastern North Atlantic): Results of single warp trawling at lower slope to abyssal soundings. Journal of the Marine Biological Association of the United Kingdom. 71: 359-373.

Merrett, N.R. & N.B. Marshall 1981. Observations on the ecology of deep-sea bottom-. living fishes collected off northwest Africa (8-27 ^oN). Progress in Oceanography. 9: 185-244.

Moore, J.A. & J.D.M. Gordon (Eds) 2003. Symposium on deep-sea fisheries. Journal of Northwest Atlantic Fishery Science. 31.

Morales-Nin, B. 2001. Mediterranean deepwater fish age determination and age validation; the state of the art. Fisheries Research. 51: 377-2383.

Morales-Nin, B., A. Canha, M. Casas, I. Figueiredo, L.S. Gordo, J.D.M. Gordon, E. Gouveia, C.G. Pineiro, S. Reis, A. Reis & S.C. Swan 2002. Intercalibration of age readings of deepwater black scabbardfish, Aphanopus carbo (Lowe, 1839). ICES Journal of Marine Science. 59: 352-364.

Morato, T. 2003. Seamounts - hotspots of marine life. ICES CIEM Newsletter. 40: 4-6.

Orlov, A.M. 2003. Impact of eddies on spatial distributions of groundfishes along waters off the northern Kuril Islands and southeastern Kamchatka (North Pacific Ocean). Indian Journal of Marine Sciences. 32: 95-113.

Perez, J.A.A., R. Wahrlich, P.R. Pezzuto, P.R. Schwingel, F.R.A. Lopes, M. & Rodrigues-Ribeiro 2003. Deep-sea fishery off southern Brazil: recent trends of the Brazilian fishing industry. Journal of Northwest Atlantic Fishery Science. 31: 1-18.

Priede, I.G. & P.M. Bagley 2000. In situ studies on deep-sea demersal fishes using autonomous unmanned lander platforms. Oceanography and Marine Biology: an Annual Review. 38: 357-392.

Priede, I.G., P.M. Bagley, A. Smith, S. Creasey & N.R. Merrett 1994. Scavenging deep demersal fishes of the Porcupine Seabight, north-east Atlantic: Observations by baited camera, trap and trawl. Journal of the Marine Biological Association of the United Kingdom. 74: 481-498.

Reis, S., D. Sena-Carvalho, J.H. Delgado & M. Afonso-Dias 2001. Historical overview of the black scabbardfish (Aphanopus carbo Lowe, 1839) fishery in Madeira Island. NAFO SCR Doc. 01/103: 6pp.

Roberts, J.M., S.M. Harvey, P.A. Lamont, J.D. Gage & J.D. Humphery 2000. Seabed photography, environmental assessment and evidence fo deepwater trawling on the continental margin west of the Hebrides. Hydrobiologia. 441: 173-183.

Rogers, A. 1994. The Biology of Seamounts. Advances in Marine Biology: 30. pp. 305-350.

Smith, D.C., G.E. Fenton, S.G. Robertson & S.A. Short 1995. Age determination and growth of orange roughy (Hoplostethus atlanticus): a comparison of annulus counts with radiometric ageing. Canadian Journal of Fisheries and Aquatic Science. 52: 391-401.

Smith, P.J., S.G. Robertson, P.L. Horn, B. Bull, O.F. Anderson, B.R. Stanton & C.S. Oke 2002. Multiple techniques for determining stock relationships between orange roughy, Hoplostethus atlanticus, fisheries in the eastern Tasman Sea. Fisheries Research. 58: 119-140.

Stene, A. & A. Buner 1991 Forsøksfiske med liner vest for Shetland, Færøyene og Hebridene, samt på Reykjanesryggen. MS. Møreforsking Ålesund, Norway (Unpublished Report): 60pp.

Sulak, K.J. & S.W. Ross 1996. Lilliputian bottom fish fauna of the Hatteras upper middle continental slope. Journal of Fish Biology. 49 Supplement A: 91-113.

Sündermann, J.E. 1986. Landolt-Bornstein numerical data and functional relationships in science and technology. New series, Group V: Geophysics and space research, 3. Oceanography. Subvolume a. Springer-Verlag, Berlin.

Swan, S.C. & J.D.M. Gordon 2001. A review of age estimation in macrourid fishes with new data on age validation of juveniles. Fisheries Research. 51: 177-195.

Swan, S.C., J.D.M. Gordon, B. Morales-Nin, T. Shimmield, T. Sawyer & A. Geffen 2003a. Otolith microchemistry of Nezumia aequalis (Pisces: Macrouridae) from widely different habitats in the Atlantic and Mediterranean. Journal of the Marine Biological Association of the United Kingdom. 83: 883-886.

Swan, S.C., J.D.M. Gordon & T. Shimmield 2003b. Preliminary investigations on the uses of otolith microchemistry for stock discrimination of the deepwater black scabbardfish (Aphanopus carbo) in the North East Atlantic. Journal of Northwest Atlantic Fishery Science. 31: 221-231.

Tokranov, A.M. & A.M. Orlov 2002. Distribution and some features of the biology of eelpouts Lycodes brunneofasciatus and L. albolineatus (Zoarcidae) in the Pacific waters off the northern Kuril Islands and southeast Kamchatka. Journal of Ichthyology. 42: 579-590.

Tracey, D.M. & P.L. Horn 1999. Background and review of ageing orange roughy (Hoplostethus atlanticus, Trachichthyidae) from New Zealand and elsewhere. New Zealand Journal of Marine and Freshwater Research. 33: 67-86.

Trenkel, V.M., N. Bailly, O. Berthelé, O. Brosseau, R. Causse, F. de Corbière, O. Dugornay, A. Ferrant, J.D.M. Gordon, D. Latrouite, D. Le Piver, B. Kergoat, P. Lorance, S. Mahévas, B. Mesnil, J.-C. Poulard, M.-J. Rochet, D. Tracey, J.-P. Vacherot, G. Veron & H. Zibrowius 2002. First results of a quantitative study of deep-sea fish on the continental slope of the Bay of Biscay: visual observations and trawling. ICES. C.M. 2002/L:18: 1-15.

Uiblein, F., P. Lorance & D. Latrouite 2002. Variation in locomotion behaviour in the northern cutthroat eel (Synaphobranchus kaupi) on the Bay of Biscay continental slope. Deep-Sea Research. 49: 1689-1703.

Uiblein, F., P. Lorance & D. Latrouite 2003. Behaviour and habitat utilisation of seven demersal fish species on the Bay of Biscay continental slope, NE Atlantic. Marine Ecology Progress Series. 257: 223-232.

Williams, A. & J.A. Koslow 1997. Species composition, biomass and vertical distribution of micronekton over the mid-slope region off southern Tasmania. Marine Biology. 130: 259- 276.

Williams, A., J.A. Koslow & P. Last 2001. Diversity, density and community structure of the demersal fish fauna of the continental slope off western Australia (20 to 35 ºS). Marine Ecology Progress Series. 212: 247-263.