Just as the sea surface temperature has an influence on the behavior of tuna for surface fisheries, food mediates the distribution of tuna within a thermal range (Blackburn 1965, 1969b; Stretta et al., 1975; Sund et al., 1981; Bard and Stretta, 1981; Stretta and Slepoukha, 1986).
The food of tuna in nature is studied by analysis of stomach contents. Studies made in the eastern tropical Atlantic, and specifically in the Gulf of Guinea, are numerous: Postel (1954, 1955, 1955a, 1963), Marchal (1959), Bane (1963), Sund and Richards (1967), Dragovich (1969), 1970), Dragovich and Potthoff (1972), Pereiro and Fernandez (1974), Valle et al., (1979 and 1979a), Borodulina (1982), Gaikov (1983), Zavala-Camin (1986).
Dragovich (1969), in his review of the literature on food of Atlantic tuna, reviews briefly the different methods used to evaluate their food by analysis of stomach contents. These are:
the numerical method. The organisms present in stomachs are counted and reported as the percentage of each organism.
percentage of frequency-of-occurrence method. This consists of calculating the percentage of the number of fish that have ingested the same entity of food as a function of the total number of fish examined.
the volumetric and/or gravimetric method. This measures the volume (by displacement) and/or weight of each entity ingested and the total volume and/or the total weight of the stomach content of each fish.
“points” method consisting of utilizing an arbitrary scale to measure the degree of a stomach's fullness.
the “nomogram” method described by Chur (1973). In order to evaluate the degree of stomach repletion, a “nomogram” is used which is defined from the relation of ingested food weight to weight of the tuna.
the Index of Relative Importance (IRI) method developed by Pinkas et al., (1971). The IRI incorporates the three units conventionally used to study the importance of ingested prey into a single index: the frequency of occurrence, the volume or the weight, and the number of individuals composing this volume or this weight. This method has since been used in the Pacific by Olson (1981) and in the Atlantic off the Brazilian coast by Ankenbrandt (1985).
“Mean Volumetric Ratio Measurement” (MVRM) method described by Ankenbrandt (1985) is a variant of the Pinkas et al., (1971) method and minimizes the importance given to the number of prey in the Pinkas index.
the “Relative Restored Mass” (RRM) method described by Borodulina (1982). The RRM index is obtained by multiplying the average weight of the stomach contents by the number of prey items found in the stomach.
220.127.116.11.1. Origin of data
Among the works cited above only those of Postel (1955 and 1963), Marchal (1959), Bane (1963), Sund and Richards (1967) Dragovich (1970), Dragovich and Potthoff (1972), Valle et al., (1979 and 1979a), Borodulina (1982), Gaikov (1983) and Zavala-camin (1986) will be used. We will analyze the stomach contents of the three principal commercial species of the Gulf of Guinea: yellowfin, skipjack and bigeye.
18.104.22.168.2 Qualitative and quantitative composition of tuna diets
Before starting a study on the composition of tuna food, most authors analyze percentages of caught tuna with empty stomachs. These percentages are highly variable which is probably a function of the conditions of fishing and stomach collection. In addition, this notion of vacuity permits certain authors to study the hour of the day or the period of the year during which tuna feed.
As Dragovich and Potthoff (1972) point out, qualitative comparisons between works of different authors are difficult because of often incomplete identification of the prey ingested by tuna. Dragovich (1969) draws the list of species in the stomachs of seven principal species of Atlantic tuna. However quantitative comparisons between these different studies generally depend on comparing principal categories of food (fish, crustaceans and cephalopods).
Table 6.11 presents the number of different forms found in the stomachs of the three principal species of tuna fished in the Gulf of Guinea (yellowfin, skipjack and bigeye) described by different authors. In his bibliographical review on food for tuna in the Atlantic, Dragovich (1969), cites more than 500 different identified forms in the stomachs of seven principal species of tuna. Out of these 500 identified forms, he finds that the diet of these species is composed of: 63% fish, 21% crustaceans, 14% mollusks 2% tunicates. Most fish ingested are adults, juveniles and larvae of pelagic species. The crustaceans are in mostly macrozooplanckton and micronection (principally shrimp). The mollusks are almost always cephalopods. The rest is represented by tunicates and at times even feathers (Postel, 1955), pieces of wood (Alverson, 1963) or mangrove seedlings (Marchal, 1959). From the list of ingested prey, Marchal (1959) separates them in several categories: surface pelagic species, bathypelagic, benthic and coastal. If Beebe (1936) notes occasionally species of benthic fish in stomachs of yellowfin fished in the Bermudas, Marchal (1959) finds that the reputedly benthic species that he finds in yellowfin stomachs, are postlarval stages or juveniles in a pelagic phase.
Table 6.11 Number of forms (genera or species) identified in the stomach contents of yellowfin, skipjack, and bigeye in the eastern tropical Atlantic and in the eastern tropical Pacific by different authors.
|Sund & Richards 1967(1)||17||10||9||6||2||1||-||-||-|
|Sund & Richards 1967(2)||12||6||4||3||1||2||-||-||-|
|Dragovich & Potthoff 1972(3)||69||27||18||61||35||9||-||-||-|
|Dragovich & Potthoff 1972(4)||20||10||6||55||11||11||-||-||-|
|Valle et al. 1979||60||15||6||-||-||-||-||-||-|
|Valle et al. 1979b||-||-||-||-||-||-||29||10||3|
F = Fish
C = Crustaceans
M = Mollusks
(1) = “GERONIMO” cruise January - May 1964
(2) = “GERONIMO” cruise July - October 1964
(3) = “UNDAUNTED” cruise February - April 1968
(4) = “UNDAUNTED” cruise September - November 1968
Table 6.12 presents the quantities of food ingested and measured volumetrically for the three principal species of tuna. The variability of results is a reflection of the period during which tuna were fished and the fishing locations. In their study, Dragovich and Potthoff (1972), analyze the stomach contents of tuna caught during two cruises of the N/O Undaunted along the coast between Nigeria and Angola, while Valle et al., (1979) study the stomach contents of tuna fished by longline over a period of sixteen months throughout the Gulf of Guinea.
We will not be able to combine in one table the different values of percentage of occurrence of the three principal groups of food found in the stomachs of tuna. This is because of the fact that most authors present tables of percentages of occurrence, not by groups, but by families or species found in the stomachs. These types of tables are published by Dragovich (1970) for yellowfin and skipjack, and by Valle et al., (1979a) for bigeye. For information only, we will present the percentages of frequencies of occurrence published by Sund and Richards (1967). Among the 171 yellowfin stomachs and the 72 skipjack stomachs analyzed, these authors note the presence of fish in 76% of yellowfin stomachs and in 73% of those of skipjack, the crustaceans are found in 53% of yellowfin stomachs and 22% in those of skipjack. Finally, one finds cephalopods in 40% of yellowfin stomachs and 14% in those of skipjack.
As for the volume of the stomach content itself, Postel (1955), notes the remarkable capacity of ingestion of yellowfin. He finds a stomach content of 1.15 kg in a 24 kg yellowfin male (4.8% of its weight). Dragovich (1970) thinks that the maximum capacity of a yellowfin or skipjack stomach reaches 7% of its weight. In regards to skipjack, Dragovich (1970) estimates that the fish is capable of ingesting up to 15% of its weight per day. These values approach those found by Kitchell et al., (1978) from studies taken from skipjack in captivity; values obtained experimentally go from 10% to 19%, depending on skipjack activity. From the principle that 1.0 ml of stomach content is equivalent to 1.0 g of ingested food, Dragovich (1970) as well as Dragovich and Potthoff (1972), find that the total volume of yellowfin and skipjack stomach content is in almost all cases under 1% of the weight of the fish. These low volumes of stomach contents may be due to long periods between two feedings, to scarcity of food and/or to the fact that most ingested organisms are small macrozooplankton organisms. These two authors associate low stomach contents volumes with rapid (or even very rapid) digestion. This rapidity of digestion has been since demonstrated by Kitchell et al., (1978).
Table 6.12 Volumes of the principal food groups found in the stomach contents (expressed in percentage of total volume) of yellowfin, skipjack and bigeye in the eastern tropical Atlantic, in the eastern tropical Pacific and in the Indian ocean by different authors.
|Sund & Richards 1967(1)||55||8||19||96||2||1||-||-||-|
|Dragovich & Potthoff 1972(2)||63*||14*||22*||80*||6*||11*||-||-||-|
|Dragovich & Potthoff 1972(3)||44*||9*||46*||58*||25*||16*||-||-||-|
|Valle et al. 1979||72.8||2.5||20.6||-||-||-||-||-||-|
|Valle et al. 1979b||-||-||-||-||-||-||65.3||2.3||32|
|Waldron & King 1963(4)||46.7||24.8||27.9||74.6||3.7||19.9||62.3||1.7||35.9|
|Thomas & Kumaran 1963(5)||72||26.3||-||-||-||-|
|Kornilova 1981 (6)||58.0||14.3||27.6||56.3||10.3||33.3|
P = fish;
M = mollusks
* Values calculated from a figure in the article considered.
(1) = “GERONIMO” cruise, January - May and July - October, 1964
(2) = “UNDAUNTED” cruise, February - April, 1968
(3) = “UNDAUNTED” cruise, September - November, 1968
(4) = Skipjack from the central Pacific (results of several authors)
(5) = Percentages expressed as a function of weight of the stomach contents.
22.214.171.124.3. Feeding behavior
Yellowfin, skipjack, bigeye as well as tuna in general, do not have a preferential food. This remark is valid at the species level. In a given zone, there is nevertheless, a certain choice as witnessed in the variations of percentage in the three groups of ingested prey (fish, crustaceans and mollusks) for fish of different sizes. Tuna feed on fish, crustaceans, and pelagic and epipelagic mollusks, including larval and juvenile forms of these groups as well as larval and juvenile forms of tuna. One could qualify them as “opportunist feeders”, feeding on all animals that move and that they can see. Vision plays an important role; Murphy (1959) finds the incapacity of albacore to catch prey in turbulent coastal waters. Magnuson (1963) notes that albacore (as all tuna) is a visual predator. One can ask with Sund et al., (1981) how tuna take dead bait on longline hooks at depths over 300 meters where light intensity is particularly weak.
126.96.36.199.4. Geographic variation in feeding behavior
Bane (1963) applies his own observations and those of Marchal (1959) emphasizing the fact that in the Gulf of Guinea, yellowfin living near the coast feed principally on live fish in schools (anchovies, sardines), while yellowfin living around the islands feed principally on triggerfish, monocanthids and invertebrates. Yellowfin in the high seas feed on cephalopods and pelagic organisms.
188.8.131.52.5. Tuna feeding and productive zones
Dragovich (1969) points out the fact that in the Atlantic, outside of Soviet work (cited by Dragovich, 1969), very little studies support the relation between tuna food and zones of productivity. Several authors attempt to link tuna distribution to that of its food: Blackburn (1968), Legand et al., (1972), Roger and Grandperrin (1976) in the Pacific and Beardsley (1969), Dufour and Stretta (1973), Herbland and Stretta (1973) in the Atlantic. The latter note that: “the link between micronecton and tuna remains conjectural because of the inability of nets to catch tuna prey, the diversity in diet, and the nonsimultaneous studies of micronecton and stomach contents.”
Dragovich (1970) is the only author to approach the problem of the food chain leading to tuna by analyzing the stomach contents of fish ingested by yellowfin and skipjack. The results of this study confirm the dependence of prey organisms towards on macrozooplankton. Copepods are prominent in the stomach bolus of ingested fish. Dufour and Stretta (1973), note that the zooplankton and micronecton are abundant in frontal type thermal structures where tuna concentrate.
184.108.40.206.6. Seasonal variation of food
Sund and Richards (1967) note a difference in the presence of some organisms in yellowfin and skipjack stomachs between cruises in warm and cold seasons by the Geronimo. From the two cruises of the N/O Undaunted carried out in both warm and cold seasons, Dragovich and Potthoff (1972) obtain comparable results. Although these four authors find differences on the whole in tuna food in warm and cold seasons, it is not the same if one examines families of ingested taxa. Sund and Richards (1967) discover dactylopterids in stomachs in warm and cold seasons, while Dragovich and Potthoff (1972) only find this family in the warm season. Certain species of crustaceans (Phronima sedentaria, Phronima semilunata, Euphausia sp.) only appear according to Sund and Richards (1967) in the warm season while Dragovich and Potthoff (1972) observe them during both seasons. For yellowfin and skipjack, more sampling is necessary in order to determine whether the presence of certain prey as a function of the hydrological season of the Gulf of Guinea is significant.
The large source of data gathered by Gaikov (1983) has permitted him to discern general features of seasonal variations of bigeye food in the Gulf of Guinea. According to this author, the degree of stomach repletion is maximum from April to August and in October-November and decreases in September.
220.127.116.11.7. Food as a function to the time of day
The diurnal variation of skipjack feeding observed by Dragovich (1970) is identical to that observed by other authors in the Pacific (Waldron, 1963; Nakamura, 1965); skipjack feed actively in the morning and before sunset, but not at night. Dragovich (1970) analyses with the help of a statistical test (for skipjack fished in the Gulf of Guinea), the average stomach content volume as a function to the time of day. The percentage of empty stomachs is higher and the average volume of the bolus is lowest near noon.
For yellowfin, Bane (1963) estimates that this species feeds principally early in the morning (before 1000 hrs) and late in the afternoon (after 1600 hrs) but is capable of feeding at any hour of the day. Talbot and Penrith (1963), note that in the high seas of South Africa, yellowfin as well as bigeye feed principally early in the morning, then feeding activity slows down during the day and increases at the end of the day. According to these two authors, yellowfin do not feed at night while bigeye are capable of so doing. For bigeye caught by longline in the Gulf of Guinea, Fedoseev and Chur (1979), from 162 tuna studied, find that bigeye show an increased index of repletion between 1400 and 1500 as well as between 1700 and 1800 hours.
18.104.22.168.8. Food in relation to tuna species and size
Dragovich (1970) compares yellowfin and skipjack food. He notes a clear similarity in diets of these two species in the Atlantic taken as a whole. In the Gulf of Guinea, he finds that out of 38 families of fish found in yellowfin stomachs and 21 for skipjack, 20 families of fish are common to both. In examining table 6.11, one notes a larger number of forms identified in yellowfin stomachs in relation to those of skipjack. This higher diversity in yellowfin food is without doubt linked to the fact that this species reaches a larger size than skipjack. However, in analyzing stomach contents of yellowfin and skipjack fished simultaneously, less than half of the taxa are common to both species: feeding by both species was selective. Dragovich and Potthoff (1972) tested by X2 the homogeneity of the relation of fish eaten to the total volume of food by yellowfin and skipjack fished in the same school. The difference between the two species is only significant for only one of the two cruises of the N/O Undaunted. Zharov et al., (1964) (article not consulted but cited by Dragovich, 1969), observes differences in yellowfin and skipjack; the former fed on a great variety of organisms from macrozooplankton to fish, while the latter consumed juvenile fish, sardines, cuttlefish and small crustaceans. The number of crustaceans eaten by skipjack is relatively larger than for other tuna such as yellowfin (Cayré, 1984).
Bane (1963) finds that the volume of the yellowfin stomach bolus increases proportionally with its size. However, he notes that stomachs of small yellowfin contain, in relation to body weight, more food than those of large yellowfin.
Dragovich and Potthoff (1972) study, for different yellowfin and skipjack size classes, the stomach contents volumes and the frequency of presence of the three principal groups eaten by tuna. From this study, he concludes that the percentage of the volume and the frequency of occurrence of fish eaten increase with tuna size; Cayré (1984) adds that this results both from better swimming performances of older skipjack and from the growth of their gill rakers that retained less prey of small sizes such as crustaceans. In conclusion, Dragovich and Potthoff (1972) estimates that there is on the whole very little difference in food found in the stomachs as a function to body weight of the two species.
Based on stomach contents of purse seine caught yellowfin in the Gulf of Guinea from the Côte d'Ivoire from December 1981 to March 1982, Zavala-Camin (1986) studied on cannibalism among tuna. As prey, skipjack have only been found in 6 yellowfin stomachs out of 218 examined. From skipjack fished by the Tema pole and line fleet, this author does not find one skipjack among the 145 skipjack examined. Relying on a bibliographical review, Zavala-Camin (1986), calculates the percentages of skipjack found in the stomach contents of predators (tuna and billfish). If one extracts from this bibliographical analysis information concerning skipjack found in tuna stomachs, one finds that for skipjack, the rate of cannibalism is very low: out of 6,226 stomachs of skipjack analyzed by nine authors in the Atlantic, Pacific and Indian Oceans, only 17 stomachs contain skipjack, a rate of cannibalism of 0.26%. For other tuna, 7.8% of stomachs of yellowfin fished in the three oceans contain skipjack; for bigeye, this rate increases to 5.2%. In this analysis, he concludes that the principal predators of skipjack are more often billfish (Istiophorus platypterus, Makaira nigricans, Makaira indica, Tetrapturus audax and Tetrapturus albidus).
22.214.171.124.9. Other types of analyses made in the Atlantic, Indian and Pacific oceans
Dragovich (1969) cites authors using tuna as collectors of marine organisms; in this way, the works of Bouxin and Legendre (1936) describing pelagic fauna from the Bay of Biscay are most notable.
Variation of food as a function of the distance to the coast. Waldron and King (1963) in analyzing the stomach contents of skipjack from the coast (central Pacific islands) to 200 miles and more off shore, finds that the stomach contents increase the further one goes from the coast up to a distance of 50 miles. Above this distance, the variations are irregular.
Variation of food as a function of the depth. These analyses are made in the Pacific (Legand et al., 1972) and in the Indian Ocean (Kornilova, 1981).
Blackburn and Laurs (1962) present, for the eastern tropical Pacific, maps of the diurnal and nocturnal skipjack prey distribution by relying in part on the works of Alverson (1963), Nakamura (1965) and Waldron and King (1963) concerning tuna food and on the “EASTROPAC” cruises during which prey organisms were sampled with a net between the surface and 200 meters.
126.96.36.199.10. Problems posed by the feeding of tuna
In the introduction, we indicated that the distribution of tuna is determined by the presence of food within a temperature interval. As Kitchell et al., (1978) point out: “it may seem paradoxical that there could be lack of food while it is a factor limiting the growth of tuna given that they live in surely the poorest area frequented by fish: the epipelagic zone of tropical seas”.
This paradox does not stop there, as Roger and Grandperrin (1976) find that tuna feed only during the day on epipelagic prey and that they feed very rarely on micronectonic fish that migrate vertically in the night; these fish constitute the major part of the D.S.L. (Deep Scattering Layer). Also, prey fish for tuna feed as well during the day on euphausids (Stylocheiron) that do not migrate. These two authors state however that all links in the food chain inhabit the layer 0–450 m in the day and feed in the day. According to these authors, this implicates that the epipelagic ichthyofauna throughout the food chain can not benefit from the considerable biomass of migratory species. This analysis must, for our purposes, be modified as tuna are opportunistic visual predators for which the food spectrum is vast. This permits them to feed on that which is the most accessible at the place where they are found (Blackburn, 1968 and Valle et al., 1979) and to feed on species which are capable them-selves of feeding at night on migratory species. According to Roger and Grandperrin (1976), tuna through cephalopods contributing a good part of their food ration, benefit somewhat from this migratory faun. Cephalopods also display a nocturnal feeding activity.
To the paradoxical situation expressed by Kitchell et al., (1978) and Roger and Grandperrin (1976), we will point out as Sund et al., (1981), that it is vital for tuna to move toward relatively rich zones in which they would be able to aggregate on food concentrations which may be found in places where the temperature is favorable.
Yellowfin, skipjack and bigeye have a very varied diet: these are active visual predators in the epipelagic zone of the ocean which hunt in the early morning and at the end of the afternoon with the possibility for bigeye, to feed at night. If all tuna have a fairly similar diet, it would seem that within a single school, yellowfin and skipjack would not always consume the same prey but would however have a preference for such fish as they would be capable of catching.
The qualitative variations of food as a function to the hydro-climate are expected because of the opportunist feeding behavior of tuna; on the other hand, the absence of study on quantitative variations of yellowfin and skipjack food as a function to season does not allow a whole picture of the biology to these two species linked to reproduction and migration. However, the works of Kitchell et al., (1978) show that the availability of food is a factor limiting the growth of yellowfin and skipjack juveniles, whereas for adult fish the limiting factor is their capacity to consume and assimilate available food.
Kitchell et al., (1978), applying on the works of Reid (1962) in the central Pacific, evaluates at 2.5 ppb the quantity of prey organisms in the sea. In spite of this theoretically very diluted food, tuna can survive, which demonstrates primarily the highly uneven distribution of food and secondly the remarkable capacity of tuna to localize and to “harvest” aggregations of prey organisms.
It is mainly American researchers who have been interested in the very exceptional phenomenon in the world of fish of a form of conservation of internal heat seen in tuna and in some sharks (Lamnidae). Although this phenomenon has not been the object of research or experiments conducted on tuna from the study zone, it is justifiable to review here the state of knowledge acquired in this field and to make a synthesis of hypotheses and real facts on this subject, so characteristic of tuna, without respect to geographic zone.
Since ancient times, there are observations according to which the flesh of recently caught tuna seems warmer than the water where they were fished. The first person to precisely describe this phenomenon was the English physician Davy who claims to have measured differences of around 10°C between the sea water temperature and of the flesh of certain tuna (Davy 1835). It was not until 1923 that the Japanese biologist Kishinouye told of this faculty of tuna to maintain an internal heat over that of their environment with the presence in these species of a very specific circulatory system that he calls the “vascular heat exchange system by counter current circulation” (or rete mirabile). This faculty to conserve heat produced by metabolism and the presence of this very particular vascular system are specific to tuna and certain sharks and are unique in the world of fish. In the family Scombridae, this heat exchange system allows the distinction of all of the 13 tuna species (group thunnini) from all other species of this family (Kishinouye, 1923; Gibbs and Collette, 1967; Collette, 1978).
The principal of this system can be schematized in the following manner: blood, heated by metabolic activity is carried by the veins toward the gills to be reoxygenated; the warm venous blood gives a part of its heat to the well oxygenated but colder blood that comes from the gills and is carried in the arteries toward the muscles; this heat exchange is made because of the existence of a web of very fine and very close capillary vessels, in which the venous capillaries transporting the “warm” blood, cross a complicated network of arterial capillaries that carry the “cold” blood. The flow of blood in these two types of capillaries is slowed by the complexity of the network. As the circulation in the two networks, arterial and venous, is in opposite directions, there is a heat exchange between the “warm” venous blood and the “cold” arterial blood.
After Kishinouye, it was not until the sixties and seventies that several authors were able to actually measure with precision on recently caught tuna, or observed directly at sea with the means of acoustic tags, the differences in temperature between the water and the blood of various species of tuna and to analyze precisely the circulatory system of these species (Barrett and Hester, 1964; Carey and Teal, 1966; Carey et al., 1971; Stevens and Fry, 1971; Stevens et al., 1974). In addition numerous works are cited which specify the intimate mechanisms permitting the conservation of heat in relation to tuna metabolism and to analyze the consequences on the behavior and distribution of the species (Neill et al., 1972; Neill and Stevens, 1974; Dizon et al., 1974; Graham, 1973, 1975; Neill et al., 1976; Dizon et al., 1976, 1977, 1978; Barkley et al., 1978; Brill, 1978; Brill et al., 1978; Graham and diener, 1978; Sharp and Vlymen, 1978; Stevens and Carey, 1981).
In this chapter we will point out the differences that exist in the various circulatory heat exchange systems and that permit the distinction of three groups of tuna species; next we will list advantages or constraints that the heat exchange system seems to imply before mentioning how it can mediate the behavior and distribution of the species.
This system, the general description of which was given in the introduction, can be found in three different areas of the body which leads to the distinction of 3 types of heat exchange:
a lateral heat exchange system or LES (figure 6.21), composed of one or two rete mirabile, localized on each side of the body and in each of which arterial and venous capillaries stemming from a cutaneous artery and a cutaneous vein intersect. The anatomy of the whole system (origin of cutaneous arteries, connection of the cutaneous veins, association of capillaries…) and the degree of its development permitting the distinction and classification (phylogeny) of the different tuna species (figure 6.21).
a central heat exchange system or CES situated under the spine in the hemal arch (figure 6.21) and where the rete mirabile is formed by the association of venous capillaries joined to the posterior cardinal vein and of arterial capillaries connecting the dorsal aorta. The extent and organization of the rete mirabile, the more or less straight association of the dorsal aorta and the posterior cardinal vein, and the more or less close position of the CES to the ventral side of vertebrae in the hemal arch, as well as the simple presence or absence of the whole central heat exchange system, are also characteristic of the different tuna species.
a visceral heat exchange system (VES) situated on the ventral side of the liver and composed of several retia mirabilia, or vascular cones (1 to 5 cones per lobe of liver), associating the blood vessels that flow on the ventral sides of the liver lobes with the dorsal sides of each of these lobes. The presence or absence of this system also allows the distribution of tuna in two distinct groups.
Figure 6.21 Transverse sections through the medial part of the body of four species of tuna and positions of the lateral (LES) and central (CES) heat exchangers. The placement of red muscle is indicated by the the stippled areas (after Graham, 1975). ad=dorsal aorta; vcp = posterior cardinal vein; vc = cutaneous veins; ac = cutaneous arteries.
The presence of a lateral heat exchange system (LES) is characteristic in all tuna species (Thunnini). The degree of development of this system allows the phylogenetical classification of the different tuna species (figure 6.21). The most primitive genera is Auxis that has only one pair of cutaneous arteries. In other species, the LES and the cutaneous vascularisation are increasingly developed when one moves on to skipjack, Katsuwonus (2 pairs of cutaneous arteries), then successively to different species of the genus Thunnus i.e. yellowfin (T. albacares), bigeye (T. obesus) and bluefin (T. Thynnus). This LES development seems to be paired with an important reduction in the central heat exchange system (Sharp and Pirages, 1978). The latter is practically absent in bigeye and nonexistent in bluefin, a species in which there is no posterior cardinal vein.
The presence or absence of a central heat exchange system has allowed the distinction of 2 groups of tuna (Gibbs and Collette, 1967):
A group with temperate affinities, or the bluefin group (T. thynnus, T. alalunga) not having CES,
A group with tropical affinities, or the yellowfin group (Auxis, Euthynnus, Katsuwonus, T. albacares).M
The bigeye (T. obesus) would be in between these two groups because of practically no CES development and presence (as with bluefin) of a visceral heat exchange system (VES).
Within the yellowfin group, the development of the CES, as well as the placement and importance of red muscles allows the distinction of the almost nonmigratory coastal species, auxids and tunny (Auxis, Euthynnus), from the widely distributed highly migratory species as yellowfin (T. albacares) and skipjack (K. pelamis); the placement of red muscles in yellowfin (in contact with body surface) allows one to say that at equal size the yellowfin is a more tropical and less cosmopolitan species than skipjack (Sharp and Pirages, 1978).
In general, with all fish, heat produced by metabolic function and carried by the blood is dissipated in the gills. With tuna, the metabolic heat is more or less conserved, or at the very least dissipated with a certain delay because of the heat exchange system. This results in an excess of muscle heat in relation to the environment. This excess of heat when measured at the moment of capture, can go up to 21.5°C in bluefin (Carey et al., 1971), up to 21° C for bigeye (Konagaya et al., 1969), 7°C for yellowfin and 11.7°C for skipjack (Barrett and Hester, 1964; Stevens and Fry, 1971).
188.8.131.52.1. Advantages linked to heat conservation
The quantity of metabolic heat produced is a function of the intensity of muscular work. It has been calculated that 80% of energy spent by a skipjack is transformed in heat; only the remaining 20% actually is used for propulsion of the animal (Webb, 1975). Consequently, the more quickly a tuna swims, the more heat it will produce. This observation has led numerous authors to make the hypothesis that the principal interest for tuna to conserve heat is to facilitate muscular work and assure a maximum maintained swimming speed (Carey et al., 1971).
Other advantages directly linked to maintaining a relatively warm temperature or to thermal inertia created by heat exchange systems have been advanced:
to allow a certain flexibility in control of a raised metabolism such as that of tuna (Hochachka et al., 1978).
to facilate the diffusion of oxygen from the myoglobin toward the mitochondria of the contracting muscle cells (Stevens and Carey, 1981)
to permit a certain independence with respect to exterior variations in temperature and thereby facilitating movements to environments of contrasting characteristics (Graham, 1975)
to increase the rate of digestion by maintaining a somewhat higher temperature in the intestines (Magnuson, 1969).
184.108.40.206.2. Thermoregulation in tuna
The different potential advantages of a raised internal temperature and their relative importance remain on the whole hypotheses to be demonstrated. Thus, the principal question yet to be elucidated is in fact directly linked to the function and role of the heat exchange systems notably in tropical tuna that live in waters where the temperature is relatively high (above 20° C).
We have seen that the principal and apparently unique role of the heat exchange system is to conserve a part of the interior body heat produced by metabolic activity (muscles), which, at certain level of activity, can result in internal temperature being higher than external temperature of the environment. In these conditions of permanent conservation of metabolic heat, the internal temperature of tuna should, in the case of intense activity maintained during a long period (frequent in tuna), reach fatal values for individuals. It is an absolute necessity for this elevation of internal temperature to be controlled, notably during activity in warm waters, by a some sort of mechanism such does not permit the temperature to rise above a certain threshold. It is this control mechanism that is designated by the term “thermoregulation” which seems logical to invoke for tunas. Thermoregulation seems to occur in tuna in two different ways: by modification of behavior and/or internal physiological modifications.
- Behavioral thermoregulation
Retaining the only role of heat exchange system so far mentioned (conservation of a part of metabolic heat), several thermoregulation mechanisms can be invoked, either separately or jointly; they will be grouped below under the term “behavioral thermoregulation” as all cause movement or voluntary activity of the individuals.
These mechanisms are:
vertical displacements of the fish that make it go alternatively from warm surface waters to colder deep water layers. Such movements have been clearly demonstrated by acoustic tagging experiments conducted on the three principal tuna species (among others), yellowfin, skipjack and bigeye (Yuen, 1970; Dizon et al., 1978; Carey and Olson, 1982; Levenez, 1982; Cayre et al., 1986)
speed modification: decreased swimming speed to decrease the quantity of heat produced; or increaded swimming speed to increase the quantity of heat dissipated to the exterior.
augmentation of the body surface by which a large part of heat can be dissipated externally (Brill et al., 1978), notably by the deployment of dorsal fins or by any other means that can increase the water turbulence around the body.
However none of these tactics of thermoregulation seem really sufficient to dissipate the large quantities of heat produced at high levels of activity notably in tropical waters (Dizon et al., 1978; Sharp and Vlymen, 1978). Moreover, some of these tactics seem irreconcilable with tuna biology that, as great pelagic migrators, are always in search of maximum swimming efficiency.
On the basis of the observation in which the heat dissipation problem is posed especially for species frequenting habitually tropical waters (auxids, tunny, yellowfin, skipjack), and only in species observed having a central heat exchange system, Sharp and Vlymen (1979) have postulated that this system would serve or participate in a process, tending rather to lower the heat than to conserve it; this process could occur, if as Sharp and Vlymen suggest, the exchange system would in fact permit accelerated conduction of blood driven to the gills, the place essential to the dissipation of heat.
- Physiological thermoregulation
Dizon, Brill and Yuen (1978), stating that tuna must at times dissipate at maximum the heat produced (at a level of heightened activity), or to the contrary, to conserve this heat, postulate the existence of true physiological thermoregulation in this species. Their hypothesis is based on experiments and direct observations made on tuna (skipjack) kept in captivity in large pools; these authors noted:
on one hand, that models predicting the internal temperature of animals observed seemed to indicate that the efficiency of the heat exchange system could vary from 17 to 47% and that this would explain that the internal temperatures observed are often lower than those predicted by models.
on the other hand, none of the models used can clearly describe the complex relations (and very variable) between the internal temperature, swimming speed and the temperature of the surrounding environment.
They admit however, that for tuna the simplest means to resolve problems of retention or of dispersion of heat according to needs, is to be able to bypass voluntarily the heat exchange system that would allow the thermoregulation of tuna. This ability to decouple the heat exchange system would permit thermoregulation in tunas.
Although the physiological mechanisms for thermoregulation remain as yet poorly elucidated on the whole, two processes seem to come into play:
The first mechanism allows the limitation of the quantity of heat produced by improving swimming efficiency. As in periods of intense activity, the white muscles come in to play. These function in a largely anaerobic mode (tending to decrease the consumption of oxygen), and produce therefore much less heat than the red muscles; also they are not linked to the circulatory heat exchange system.
According to the second mechanism, again largely hypothetical, the heat exchange system would permit, in certain cases, larger quantities of heat to dissipate at the gills.
Although the role of vascular heat exchange systems, and especially their implication for thermoregulation in tuna, remains yet to be clarified, it is undeniable that the structure and the development of these systems, associated to the placement of red muscles, have direct implications on the ecology of tuna.
Several examples illustrate this role of the heat exchange system on behavior:
For sport fishermen, the nature or the development of the heat exchange system in relation to the temperature of the environment at the moment of capture seems to directly influence the combativity of the different tuna species; the bluefin, with its more sophisticated heat exchange system, would always be more combative than a yellowfin or bigeye, for example. Cases where fish are caught after an intense fight and taken aboard dead with abnormally soft flesh, can be explained by the inability of these individuals to accomplish sufficient thermoregulation in extreme and abnormal conditions of activity. Tuna presenting this soft flesh are occasionally observed in large purse seine catches, and are designated by the term “burnt tuna”. These individuals, rarely accepted by canneries, are totally unacceptable for raw consumption as “sashimi”, much appreciated by the Japanese. It is notable that in the Japanese market, the species of tuna which have the highest commercial value are precisely the species that have the most efficient heat exchange system (bluefin, bigeye) and in which “burnt tuna” is rarely observed.
The different characteristics of the heat exchange system explain, to a large degree, the ecological affinities (in regard to environmental temperature) of various tuna species or, in the same species, different size ranges of the species in question. The following separations can be made:
small coastal tuna with a strictly tropical affinity: Auxis (auxids), Euthynnus (tunny) …, species where the central heat exchange system is well developed, the lateral heat exchange system barely functional, and which posses large red muscles in surface contact with the exterior.
young individuals (FL < 70 cm) with a tropical affinity quasi-exclusive of the species T. albacares (yellowfin) and T. obsesus (bigeye) where the heat exchange systems are poorly developed in relation to those of adults.
adult yellowfin with a tropical affinity, but less marked than that of the young, in the sense that they are frequently observed at horizontal and vertical limits of the species habitat in waters of 15°C to 20°C.
cosmopolitan species like skipjack (K. pelamis), where the distribution area extends largely beyond the only tropical regions.
adult bigeye with a rather temperate affinity 10°C – 18°C, (contrary to young ones) and that are more often seen in cold (10 – 15°C) deep (300 – 400 m) waters in strictly tropical regions, or in those of the surface of the north and south regions of the species habitat.
tuna with a fairly rigorous temperate affinity (12 – 18°C) as albacore (T. alalunga), or temperate but very cosmopolitan (5 – 28°C) as bluefin (T. thynnus) are the most advanced phylogenetically tuna species.
The importance and interest in understanding the phenomenon of thermoregulation in tuna is clear since it directly influences the vertical and horizontal distribution (biogeography) of the different species.
This knowledge of thermoregulation of the different species is completely transposable to tuna of the Gulf of Guinea, and allows the understanding of certain characteristics of tropical tuna fisheries such as the apparent preferential abundance zones and the catchability of the different species (yellowfin, skipjack and bigeye) or, within a given species, size ranges of exploited individuals.
The direct correlations between certain abiotic factors and species distribution estimated from catches has been noted for many years. The most frequently cited factors or parameters, essentially because their ease for of measurement at sea, are the temperature, salinity and dissolved oxygen. A number of correlations have been and are still calculated between these parameters and certain variables of a biological character. It is necessary however to recall here that a correlation between two variables does not imply a priori the existence of a direct cause and effect relationship between the variables.
We have seen (paragraph 6.1.5.) that the water temperature could, under certain variable conditions, depending on species and stage of development, be a factor that controls the distribution of tuna in the ocean, i.e. that limits in certain cases (extreme values, larval stages) this distribution.
On an oceanic scale, salinity itself does not seem to play a determining role in the distribution of tuna. Exceptions must be made, however, for certain specific, local situations in which extreme and abnormal values of salinity can constitute a limiting factor. In the Bay of Biafra, there is a strong local desalinization created by the outlet to sea of large fresh water flows which could explain to the great scarcity of tuna in this region.
As for oxygen, the life and development of fish are based on utilization of this element. Oxygen is present in sea water in a dissolved form; it is extracted and absorbed by the blood at the gills, and carbon dioxide is released. Oxygen is a vital element for fish in general, and for tuna particularly because of their very high metabolism. Several adaptations of the circulatory system of tuna gills and of specific modalities of respiration in relation to swimming give to these species an exceptionally high dissolved oxygen extraction capacity. The efficiency in dissolved oxygen extraction can rise to 90% (Stevens, 1972), while it is only 30% in most other teleost species. There is great interest in knowing and understanding in how the amount of dissolved oxygen in the ocean can constitute a limiting factor on survival and distribution of different tuna species.
Studies of this type assume an essentially physiological and ethological character; they require by their nature and in order to take into account constraints linked to size and pelagic behavior of the species considered, major material facilities (various equipment, experimental basins…). For these various reasons, these experiments have been mainly carried out in the United States and numerous references to these works are made in G.D. Sharp and A.E. Dizon (1978). In order to not make a detailed synthesis here of these numerous, often very specialised works, we will limit ourselves to mention the results concerning oxygen needs of the three principal species of tropical tuna (yellowfin, skipjack, bigeye) and we will see next how catches of these species are distributed as a function of average concentrations of dissolved oxygen observed in the Atlantic Ocean.
After demonstrating the fact that the respiration rate (oxygen consumption) of tuna in general is exceptionally high, and from 2 to 10 times that of other teleost fish in comparable activity conditions, it is proper to verify that levels of oxygen met at sea could effectively constitute a factor limiting the presence of tuna. To this effect, a certain number of experiments have been made either on certain tissues removed from tuna, or on living individuals, mainly skipjack, in order to verify how different rates of dissolved oxygen could affect these tissues or the behavior and survival of individuals. In the same way several mathematical formulations have been proposed to calculate the oxygen consumption at different levels of activity and for various tuna species and sizes of individuals (Gordon, 1968; Kitchell et al., 1974; Sharp and Francis, 1976; Neill et al., 1976; Dizon, 1977; Dizon et al., 1977; Sharp, 1978; Brill, 1979; Gooding et al., 1981). From all of these works, it is concluded as a certain fact that the three species (yellowfin, skipjack and bigeye) are sensitive to various degrees to dissolved oxygen concentrations below certain thresholds. This extreme sensitivity has been directly observed in skipjack kept in captivity (Dizon, 1977; Gooding et al., 1981); it has been verified that rates of dissolved oxygen under 3.5 mg 02/1 provoke accelerated swimming in skipjack; this acceleration has been interpreted as flight behavior in regards to an environment unsuitable for a normal life for the species.
Although direct observations, such as the one described above, could not be made on the two other species (yellowfin and bigeye), minimum levels of oxygen specific to the survival of these species at various sizes were able to be calculated (Sharp, 1978) and are mentioned in the table below:
(FL in cm)
|Tolerance - lower limit of oxygen concentration|
(ml/O2/1 of water)
Skipjack (K. pelamis)
Bigeye (T. obesus)
Theoretically, if one compares values of tolerance thresholds mentioned above with average concentrations in dissolved oxygen observed in the Atlantic at different depths (Merle, 1978), one notes that:
levels of dissolved oxygen in the surface layer between 0 and 50 meters in depth (figure 3.12), are almost always above the tolerance thresholds of the three species. The concentration in dissolved oxygen in the Atlantic in this surface layer does not constitute a limiting factor to the distribution of the three species (yellowfin, skipjack, bigeye) in this layer of water.
from 150 meters, there are (figure 6.22) two vast zones of low oxygen concentration (< 2.4 ml/l). These two zones are localized in the eastern part of the Atlantic and centered respectively on the latitudes between 6°N and 16°N and between 6°S and 18°S. In theory, in these zones and at these depths, yellowfin should be rare, skipjack absent and only bigeye could subsist there. When one considers deeper layers (figure 3.13), the surface of the the zones “poor” in dissolved oxygen is more and more extended as depth increases. The concentration of dissolved oxygen is lower and lower to a depth of 500 meters, from which it begins to increase.
Practically, it is interesting to compare the geographical distribution of catches of the three species with theoretic considerations shown above.
Skipjack: it is known from the distribution of surface tuna (pole and line boats and purse seiners) catches that this species is mainly limited to the intertropical zone. The extension of its distribution in the surface layer in the more northern or southern regions would be controlled by the temperature rather than by levels of dissolved oxygen (Cayre, 1984). The distribution at depth of this species is difficult to determine even by examination of longliner catches, the only gear to actually exploit layers of water between 100 and 300 meters in depth. Indeed, although long line skipjack catches exist, notably in the central Atlantic (Pianet and Yanez, 1979), they are very rare, as the species is not often sought by fishermen using this gear. It is however clearly demonstrated that skipjack can not subsist normal in waters where the concentration in dissolved oxygen is under 3 ml/l, even if very brief incursions in depth notably in waters low in oxygen concentration have at times been observed during acoustic tagging experiments (Levenez, 1982).
Yellowfin and bigeye: We will not dwell on the surface distribution of these species in relation to oxygen concentration. As for skipjack, oxygen never reaches levels below those corresponding to tolerance thresholds of these two species in the water layer considered (0–50 meters).
Figure 6.22 Mean annual distribution of the concentration of dissolved oxygen (ml/l) in the Atlantic at a depth of 150 meters. Note that the relative maximum concentration is located in the equatorial region, while the minima concentration are found off the coasts of Senegal and Guinea and off Angola (after Merle, 1978). For comparison, refer to figures 3.12 and 3.13 where the distributions of oxygen concentration at 50 m and 200 m are presented.
The north and south extension of yellowfin distribution is perhaps controlled by other factors, notably the water temperature (paragraph 6.1.5.). As for bigeye, one knows that is largely widespread up to high latitudes (45° N - 45 S) because of its great tolerance to low temperatures (paragraph 6.1.5.).
In order to completely understand the extension of the habitat of these two species, one can not be limited to the only catch distribution made by surface gears (pole and line, purse seine…) for which the efficiency is limited by definition of the water layer between the surface and a hundred meters in depth. On the other hand, longlines which have been in widespread use for nearly thirty years throughout the Atlantic, including central regions, permits the exploitation of deeper water layers. Two types of longlines are actually used in the Atlantic to capture tuna, marlin and swordfish at depth:
the so-called “classical” longline, the only type of longline employed in the Atlantic up to 1975, allows the exploitation of the water layer roughly between 70 and 180 meters in depth.
the “deep” longline, which was introduced in the Atlantic in 1975 by Japanese fishermen, allows the exploitation of the warmer layer between 70 and 300 meters in depth (Suzuki and Kume, 1982).
Examination of the geographical distribution of yellowfin and bigeye catch rates achieved from 1956 to 1983 by longliners (figures 6.23 and 6.24) shows a certain number of fundamental differences in the distribution of the two species:
yellowfin catches are essentially limited to equatorial and sub-equatorial zones
those of bigeye, although also observed in the equatorial region, are particularly large in two regions in the eastern part of the Atlantic off Senegal (between 10° N and 20° N) and off Angola (between 10° S and 20° S) respectively.
Figure 6.23 Map of the abundance of yellowfin estimated by catch per unit of effort (or catch rate) of longliners operating in the Atlantic from 1956 to 1980.
Moreover, according to a certain number of observations (experimental fishing) made essentially in zones where bigeye is abundant (north-east Atlantic), there also seems to be a notable difference in the bathymetric distribution of the two species:
yellowfin become more and more rare as depth increases; they are almost absent below 200 meters in depth.
conversely the density of bigeye increases with depth and becomes maximum between 200 and 300 meters in depth.
This observation has been made empirically since 1975 by Japanese fishermen that preferentially catch bigeye, a much more appreciated species than yellowfin on the Japanese market. It was around 1975–1976 that the Japanese fishermen began to use the deep longline in the Atlantic. As mentioned above, this gear allows fishing to around 300 meters in depth, in place of the 180 meters reached by classical longlines (Suzuki and Kume, 1982), and therefore targets bigeye.
If one approaches these different observations, of distribution of dissolved oxygen levels on one hand, and the value of tolerance thresholds respective of the two species to this rate of dissolved oxygen on the other, the hypothesis that oxygen would be a limiting factor in the distribution of tuna seems to follow logically. Longline catches of yellowfin, the species in which the low oxygen tolerance threshold is around 2 ml/l, are essentially limited to zones in which the rate of dissolved oxygen is over that value and notably to the equatorial region where the concentration of dissolved oxygen is maximum (figures 6.22 and 6.23). Conversely, bigeye, less sensitive to low levels of oxygen (low oxygen tolerance threshold equal to 0.5 ml/l), even though present in the “yellowfin zones”, are found in maximum abundance in regions situated off Senegal and Angola where levels of dissolved oxygen are minimal (figures 6.22 and 6.24). A similar phenomenon may explain the differences observed in the extension in depth of the two species; bigeye, already abundant in zones poor in oxygen (Senegal and Angola) increase in abundance with depth and the correlated decrease in dissolved oxygen levels.
Figure 6.24 Map of the abundance of bigeye estimated by catch per unit of effort (or yield) of longliners operating in the Atlantic from 1956 to 1980.
This maximum abundance of bigeye in zones with a low dissolved oxygen concentration could be explained in the first analysis by the hypothesis according to which bigeye and certain billfish, being among the few species able to support these low concentrations of oxygen, are not found in competition with other species colonizing these oceanic regions poor in dissolved oxygen. As seems to be shown in acoustic tagging experiments (Carey and Robinson, 1981), the swordfish (Xiphias gladius) was capable, like bigeye, to maintain themselves for long periods of time in an environment with low concentration of dissolved oxygen. In this regard, the relative similarity in distribution of longline catch rates of bigeye (figure 6.24) and swordfish, Xiphias gladius (figure 6.25), notably in zones with a low concentration of dissolved oxygen, seems to support this hypothesis.
The demonstration of dissolved oxygen concentration tolerance thresholds specific to the three species (yellowfin, skipjack, bigeye) and the existence in the Atlantic of oxygen concentrations at times inferior to these thresholds, seems to indicate that dissolved oxygen is very much a factor which can directly limit the tuna habitat. The empirical observations made by means of the catch distribution confirms in a large part the theories deduced from laboratory experiments. One must, however, keep in mind that the tolerance thresholds mentioned do not constitute an absolute barrier to the passage of the species considered. As shown from acoustic tagging experiments, short duration incursions in an environment in theory unsuitable to the survival of individuals, are quite possible. On the other hand, if the oxygen tension constitutes in certain cases the factor limiting the habitat, it is also necessary, as we have seen, to call upon other factors to explain the limits of this habitat. The habitat, or the ecological niche occupied by a species, is in fact determined by a complex combination of multiple abiotic factors (temperature, oxygen etc…) and biological factors (availability of food, competition with other species…).
Figure 6.25 Map of the abundance of swordfish (Xiphias gladius) estimated by catch per unit of effort (or yield) of longliners operating in the Atlantic from 1956 to 1980.
Finally, in regards to oxygen only, numerous points still remain to be studied, such as knowing whether the absolute concentration of oxygen of the environment is more important for a fish than the tension of dissolved oxygen in relation to the theoretical value of the dissolution of this gas at saturation. This last value depends directly on the temperature to which it is inversely proportional. On the other hand, it is probable that the oxygen requirements vary with the concentration of red corpuscles in the blood and therefore depend on multiple parameters regulating this concentration (temperature, salinity, sex, stage of sexual development, growth…).