(i) Preamble

The teleosts are the single largest major lineages of fishes. There are about 450 families and over 24,000 named species. They first appeared in the fossil record in the Jurassic Period (some 170 million years ago), but they did not achieve high levels of diversity until later in the Cretaceous (about 70 mya).

We shall define the major sublineages (superorders) with the teleosts and explore some aspects of biology of some of the better known groups.

(ii) Definition of the teleosts (see Rosen 1982)

The teleosts are recognizable as a natural group of bony fishes based on characters in the skull and caudal skeleton.

(1) The name teleosts means "end bones" which refers to a feature of the caudal skeleton of teleost fishes. Some of the most posterior neural spines have been modified elongate bones known as "uroneurals" (see figure). These bones likely function to help support the fin rays of the largely homoceral caudal fin. The neural spine of the last vertebra has also become elongate to form the "urostyle". The "butressing" of the caudal skeleton via the development of uroneurals and fused hypural plates was probably of great inportance to the development of some of the diverse swimming modes of teleosts (e.g., from rapid "tight" bursts of thrust from the caudal area in tuna[ "carangiform swimming", to rapid ocillation of the caudal fin in boxfishes, "ostraciform swimming").

(2) Presence of a supramaxillary bone(s) (see the skull figures in the "Actinops" lectures).

(3) A vomer that is unpaired. The vomer has become fused into a single bone (the vomer is a pair of bones in earlier bony fishes such as Amia).

(4) The premaxillary bone bears most of the teeth in the upper jaw and possesses an ascending process. In addition, the ascending process interacts (via ligaments) with the ethmoid region and the pre-maxillary loses it tight connection with the maxillary bone.

(5) The lower jaw articulates with the quadrate bone only (not with the symplectic as well as in Amia).

(6) The function of the gas bladder becomes largely one of bouyancy reguation only as opposed to its use as an air-breathing organ in many earlier Osteichthyes (e.g. Amia, lungfishes, gars, bichirs).

(7) There are characteristics features of the placement of paired fins and numbers and arrangement of fin spines (see figure). In general, the pelvic fins move from the abdominal to thoracic to the jugular position and the pectoral fins move from a ventral position to up the side of the body wall. Further, spines are found in paired and median fins, and the premaxillary bone dominates the gape in higher teleosts.

The trends in upper jaw structure are shown below (1=most ancestral, 5=most advanced).

Above are the trends in postion of pectoral and pelvic fins (P1=ancestral, P4=advanced) and spines (S1=ancestral, S4=advanced).

(iii) Composition of the Teleostei:

The teleosts consist of four major lineages, three of which denote superorders (the fourth is a mixture of several superorders):

(1) Osteoglossomorpha: bonytongues; about 220 species

(2) Elopomorpha: eels and tarpons/bonefishes; about 800 species

(3) Clupeomorpha: herrings, sardines, shads; about 450 species

(4) Euteleosts: all other teleosts; about 22,000 species

(A) Osteoglossomorpha:

The Osteoglossomorpha are the most ancestral lineage of teleosts. Most are tropical to subtropical and found on all continents except Antarctica and Europe. They first appeared in the fossil record during the Jurassic. All are freshwater tropical to subtropical species. One genus, however, extends into Canada - the genus Hiodon with two species: the goldeye (Hiodon alosoides) and the mooneye (H. tergesis). These two species are part of the "Mississippi River fauna" that recolonized Canada postglacially. The goldeye extends to the upper Mackenzie River and the lower Peace River and likely used dispersal routes in Lake Agassiz to disperse here.

The osteoglossomorphs are distinguished by an unusual bite, i.e. the primary biting mechanism is between the parasphenoid (along the roof of the mouth) and the tongue, both of which are "well-tooth" hence the name "bonytongues" (see figure).

The bonytongues also include one of the largest freshwater fish, the "aprapaima" which can grow to almost 5 m in length. The osteoglossomorphs also include the elephantfishes (Mormyridae) which are the most speciose family (about 200 species), are found in central Africa, and have well developed electrocommunication abilities.

(B) Elopomorpha

The elopomorphs are a large (800 spp.) and diverse group of teleosts, many of which look completely unrelated as adults, but they all share a unique larval form known as "leptocephalus" larvae. To see pictures of such larvae click HERE .

Leptocephalus larvae are planktonic, very thin, transparent, and "leaflike" in shape. They may reach almost 2m in length in some species and the transformed juvenile may actually be shorter than the earlier larval stage. This larval stage results in extensive dispersal of larvae in the marine environment.

All elopomorphs are marine, brackish water, or catadromous (spawn in saltwater, migrate as juveniles to freshwater where they grow to maturity before returning to saltwater to spawn).

Some of the more notable groups include the tarpons and bonefishes. The Atlantic taropn, Megalops atlanticus, can reach 2.5 m in length and inhabits nearshore marine tropical and substropical habitats where its "manic" leaping abilities make it a legendary sportfish.

A school of tarpon "breaching" off the coast of Florida. Fish on!

The "true eels" (order Anguilliformes) are the largest members of the Elopomorpha (about 700 species) and possess an elongate body. They are also characterized by a lack of a pelvic girdle and lack of pectoral fins. This morphology (which several other unrelated groups of fishes have converged upon) makes it easy for these fishes to move in an about areas like coral reefs and for burrowing into the substrate where specific families of eels often dominate the fish fauna (e.g. moray eels on reefs and congrid (garden) eels). Here is an interesting article on a (again!) recently-described and novel feeding adaptation in teleost fishes.

Two species (of 15 total) are freshwater eels in the family Anguillidae.

The North American species (Anguilla rostrata) is found from Florida to the Maritime Provinces of Canada where it exhibits a catadromous life history. Adults and subadults are important predators in lakes and streams in eastern Canada. The European analogue is the European eel, Anguilla anguilla, found from North Africa to Scandinavia. After reaching maturity in lakes and rivers, adult eels migrate downstream and enter the ocean where they may migrate up to 5600 km to the Sargasso Sea area. Here adults from Europe and North America spawn and die. Leptocephalus larvae then drift in ocean currents for 1-3 years before transforming to "elvers" that then ascend local rivers, streams, and enter lakes where they grow to maturity.

In both areas, the species are important to commerical and recreational fisheries and there has been considerable debate about their status as distinct species. The only morphological feature that distinguishes them is a difference in average vertebral counts (103-111 in North America; 110-119 in Europe). The forms also show marked differences in allele frequencies at one allozyme locus (Mdh-2), but selection has been implicated as the driving factor for this difference.

Avise et al. (1986), however, found marked differences in mtDNA between the two forms of eels (about 3.5% divergent in sequence) despite the fact that populations within North America and within Europe were almost identical (near 0% sequence divergence). They concluded that the two forms were almost certainly not part of a single breeding pool in the Sargasso Sea, but how they assort duering mating is not known (i.e. do they spawn in different areas of the Sargasso Sea or are there innate behavioural differences in mate choice?). The two species appear to form a hybrid zone in intermediate locations such as Iceland (see Avise et al. 1990).

Regrettably, recent declines in population sizes of freshwater eels are a major conservation concern (see Stone 2003).

(C) Clupeomorpha

The clupeomorphs are the herrings, shads, sardines, and anchovies. They are not particularly diverse (about 450 species), but they can reach incredible abundances as they form large near surface schools in nearshore marine waters or in some freshwaters. Their proximity to coastal areas, huge abundances, and ease of conversion to products like fish meal means that they are incredibly important commercially. They originated in the Cretaceous Period and are well defined morphologically by:

(1) an intercranial connection between the swim bladder and the inner ear. The swim bladder divides into two parts anteriorly which penetrate large vesicles within the pterotic and pro-otic bones of the skull. This connection is thought to be advantageous for sound perception in open water habitats. This is known as an otophysic connection.

(2) temporal and auditory foramina (holes) and fossa (depressions or pits) in the skull (see figures)

(3) hypural plates in the caudal skeleton with the second plate not fused to urostyle (it is "free floating").

Most of the clupeomorphs are marine (about 80% of species). The three best known families are the Chirocentridae (wolf-herrings), Engraulidae (anchovies), and the herrings, sardines, sprats, and shads (Clupeidae). Almost all are zooplanktivorous, but a few are larger bodied and piscivorous. The wolf-herrings can reach 1 m in length (most clupeomorphs are only 10-30 cm) and have large, fang-like teeth and are piscivorous.

Boom and bust fisheries in the Clupeomorphs

Most clupeomorphs are found in nearshore marine or freshwaters, and some are anadromous (e.g. the shads) (see figure) and they form the basis of important commercial fisheries.

Distribution of the family Clupeidae

The Peruvian anchovetta usually leads the world in terms of tonnage of annual catch and the California sardine was the basis of a substantial fishery in California and provided much of the inspiration for Steinbeck's Cannery Row.

One of the notable features of such fisheries is that they show tremendous variation in productivity and yield; a few years of large catches are followed by dramatic declines before (hopefully) populations build up again to sustain large catches. The figure below shows just such a "boom and bust" cycle for the Japanese sardine, Sardinops melanosticta (top panel).

Such population crashes are not simply a response to overharvesting (although that clearly can and does play a role in some cases). Clupeoid fishes have gone through such collapses in the absence of fishery pressure. For instance, collections of fish scales from marine sediments (lower panel of figure) clearly indicate large fluctuations in the marine environment over the last 2000 years.

Clearly, understanding the biological bases for such population fluctuations would be interesting in its own right, but also would be of great benefit to the fishing industry.

Obtaining answers to this puzzle has been a large focus of much fisheries research over the last 100 yr. Possible causes of such fluctuations fall into the following possible mechanisms:

1. overfishing (lost of evidence for that!)

2. reproductive failure driven by environmental change

3. changes in migration patterns

4. changes in species dominance (owing to 1-3 above possibly)

5. All of the above acting together!

The figure below shows major oceanographic upwelling regions and associated fish species that depend on them. Clearly, an effects on the physical/chemical nature of these upwelling zones (e.g., El Nino events) could have dramatic effects on populations that depend on them.

What drives the population cycles?

One of the basic ideas behind what might drive these population cycles is the concept of the "critical period" in marine fishes. Hjort in 1913 first coined this idea to try and explain variation in the "Great Fisheries of Europe".

Essentially, the idea is encapsulated in the figure below. If one samples a cohort (those individuals born at a particular time) across time and determines mortality (numbers at time t, numbers at time t + 1) it is usually the case that the greatest mortality occurs at the very early life history stages. One possible reason for the high larval mortality rate is that the transition from yolk nutrition to exogenous feeding sources is a "critical" period in life. If the larvae run out of yolk resources and there is no food (e.g. plankters) available then they will quickly starve to death or be subject to sublethal nutritional effects (grow so slowly that they remain vulnerable to predators for longer). Consequently, the "match/mistmatch" of larval feeding to food resource availability may represent a time period (the critical period) where the size of the adult population (that vulnerable to fisheries) is determined.

Clearly then, environmental factors that may influence the matching of plankter abundance to the larval transition to exogenous feeding is an important factor that may drive yearly variation in population size.

Basic variations on this theme fall into density-independent (environment driven) and density-dependent (biotic driven) mechanisms:

Density-independent:

1. Match-mismatch hypothesis: essentially as outlined above. Developing larvae must "match" the production of food items in the environment during the transition to exogenous feeding. If "mismatch" occurs, mass starvation could occur. Environmental parameters (temperature, winds, currents, etc) could influence the "match/mistmatch".

2. Stability hypothesis: Again, this is food production related. Local stability in water masses, upwelling, etc, could influence food production and its spatial distribution. If the distribution of food and larvae don't overlap - trouble for larvae!

3. Variable niche hypothesis: Essentially, the environment is variable. If particular species cannot respond to changes in the environment in ecological time, they could get displaced by competitors that are better able to do so.

4. Member-vagrant hypothesis: sexually-reporducing organisms must find mates. Therefore, temporal persistence of populations is based, to some degree, on the ability of maturing individuals to aggregate to specific spawning sites (e.g. "homing") to enable life cycle "closure". The population size is governed, to some extent, by the geographic area over which a species can complete its life cycle. Population fluctuation may be driven, therefore, by losses to the local population by the failure of indivuduals to complete their life cycle within a particular area (i.e. they are the "vagrants"). Again, environmental factors (temperature) may influence homing ability and lead to vagrancy and lower reproductive potential.

Density-dependent:

Density-dependent mechanisms include intraspecific and interspecific processes. For instance, at very high densities there may be downward pressure on the population from resource competition (starvation), cannabalism, predator attraction, or low spawning efficiency. By contrast, at the low end of the cycle, fish may respond by exhibiting increases in growth rate (owing to lower intraspecific competition), lower mortality from predation (as predators switch to more abundant prey), or increased fecundity. These can all lead to rapid increases in population size.

As usual, all these factors can interact and drive population cycles which can lead to frustration when research is directed towards determining the effect of any one process.

For the most part, the best evidence is for environmenal and density-independent factors as being associated (causing?) population fluctuations in clupeoid fishes. Some references that deal with these issues are (see below): Lasker (1985) Purcell and Groves (1990), and Stephenson and Kornfield (1990).

(D) Euteleostei

The euteleosts are the largest lineage within the teleosts and comprise some 375 families and over 18,000 species. There are four major radiations within the euteleosts usually recognized at the level of superorders:

(1) Ostariophysi: about 6,000 species (dominate freshwaters of the world)

(2) Protacanthopterygii: 300 species

(3) Paracanthopterygii: 1,200 species

(4) Acanthopterygii: 13,000 species (dominate marine environments)

We'll discuss these in the next two lectures. Before we move on, a final point about the first three lineages of teleosts. Although they are less diverse taxonomically then the euteleosts, the osteoglossomorphs, elopomopomorphs, clupeomorphs are "successful"in other ways. The bonytongues might be regarded as relicts of a very successful and diverse early diversification of teleosts that have left specialized remnants on all continents (think of the African mormyrids with their keen electrocommunication abilities). By contrast, the elopmorphs have "perfected" one particular body morphology (elongate with reduced paired fins) to adopt a diversity of life styles from dominant predators on coral reefs to highly migratory catadromous fishes. Finally, the clupeomorphs include fishes that dominate (numerically and ecologically) much of the pelagic nearshore and open ocean habitats where they have evolved highly specialized schooling, planktivorous life styles.

References:

Avise, J.C., Helfman, G.S., and Hales, L.S. 1986. Mitochondrial DNA differentiation in North Atlantic eels: population genetic consequences of an unusal life history pattern. Proc. Nat. Acad. Sci. USA 83: 4350-4354.

Avise,J.C., et al. 1990. The evolutionary genetic status of Icelandic eels. Evolution 44: 1254-1262.

Bergeron JP. 2000. Effect of strong winds on the nutritional condition of anchovy (Engraulis encrasicolus L.) larvae in the Bay of Biscay, Northeast Atlantic, as inferred from an early field application of the DNA/C index. ICES Journal of Marine Science. 57(2):249-255.

Lasker, R. 1985. what limits clupeoid production? Can. J. Fish. Aquat. Sci. 42(Suppl. 1): 31-38.

Rosen, D.E. 1982. Teleostean interrelationships, morphological function and evolutionary inference. American Zoologist 22: 261-273.

Rothschild BJ. 2000. Fish stocks and recruitment: the past thirty years. ICES Journal of Marine Science. 57(2):191-201.

Sanchez-Velaso L. Shirasago B. Cisneros-Mata MA. Avalos-Garcia C. Spatial distribution of small pelagic fish larvae in the Gulf of California and its relation to the El Nino 1997-1998. Journal of Plankton Research. 22(8):1611-1618.

Stone, R. 2003. Freshwater eels are slip-sliding away. Science 302: 221-222.