Nearly all bony reef fishes have a bipartite life history consisting of a reef associated adult period and a pelagic larval period spent in open water. The length of time spent in the pelagic zone, referred to as the pelagic larval duration (PLD) varies between and within species and can be as short as a week or as long as a few months. The duration of the PLD and the events occurring during that time can have a tremendous effect on reef fishes (Bay et al. 2006; Leis 1991, 2006; Leis and McCormick 2002).
During the pelagic stage, larvae may disperse from their natal reef and thus affect the biogeographical distributions and diversity of fish species. Survival in the larval stage heavily influences recruitment rates and population sizes (Hjort 1914), and the individual condition of juveniles and the amount of post-recruitment growth is determined during this stage (Bay et al. 2006). Due to the potential for dispersal in the pelagic stage, genetic and demographic connectivity are intricately linked with what occurs to the larvae in open water. Researchers are often interested in the evolutionary and biogeographic significant genetic connectivity, while managers of fisheries and marine protected areas depend on the scale of demographic connectivity. As a result of its importance in affecting so many processes on the reef, a solid knowledge of the pelagic phase of reef fish life cycles is crucial to understanding and caring for reef ecosystems (Leis 2006).
Pelagic reef fish larvae have been known for over 100 years. For example, the highly specialized larvae of chaetodontids, called tholichthys, were first described in the late 19th century (Leis 1989). At the beginning of the 20th century scientists began to appreciate the importance of the pelagic larval. Fisherman had long noted annual variations in catches and fish abundance. These fluctuations were thought to be a result of migrations of adult fishes (Dower et al. 1997). Johan Hjort (1914) first proposed that the annual fluctuations in the abundance of commercially important species in Europe were due to variation in the recruitment success of juvenile fishes into the population and not mass migrations. Hjort (1914) suggested that larval mortality and, subsequently, recruitment success were due to sufficient food being present during the “critical period” when larvae exhausted their yolk sacs and began feeding on external sources. Thus starvation mortality was thought to be the main factor in controlling population abundances.
Hjort (1914) proposed that another factor controlling mortality, but to a lesser degree than starvation, was the transport of larvae away from suitable juvenile habitat. This suggestion was based on the assumption that larvae drift passively on the currents. The belief that fish larvae did not have the ability to influence their position in the water and that their behavior could be regarded as irrelevant grew into what has been termed “the simplifying assumption” (Leis 2006; Leis and Carson-Ewart 1998). The tenets of the simplifying assumption dictated that 1) larvae are poor swimmers, 2) the pelagic larval duration is the only biological variable influencing dispersal, 3) all larvae behave the same, and 4) once competent, larvae settle on the first suitable habitat they chance upon after drifting on the currents. It was not until the 1980s and 1990s that the tenets of the simplifying assumption in tropical species began to be evaluated and found wanting (Leis 2006).
Detailed studies on commercially important larval fishes in freshwater and temperate marine climates began in the 1950s and have outnumbered studies on tropical species ever since. With the lopsided knowledge of tropical and temperate species it was natural to apply what had been learned about temperate species to tropical species. Although some general assumptions may still be valid, as our knowledge about tropical species has grown, reasons to doubt the accuracy of assumptions based on temperate species have developed (Leis 2006; Leis and McCormick 2002).
The majority of temperate and tropical species belong to distinctly different taxonomic orders. The few species of fishes that have attracted the most attention in temperate waters belong to the orders Clupeiformes, Gadiformes and Pleuronectiformes. While the majority of tropical reef fishes belong to the order Perciformes. These orders have been separate lineages for at least 50-60 million years. Therefore, it is unsound to believe that the behavioral or morphological development of perciforms should be identical to clupeiform, gadiform and pleuronectiform fishes. To do so would be equivalent to believing that data on rodents would apply to primates (Leis 2006; Leis and McCormick 2002).
In addition to taxonomic differences, differences in adult habitat make assumptions based on temperate studies questionable. Tropical reef fish are demersal as adults and live on relatively small areas of hard bottom substrate on the continental shelf. The temperate species which are demersal live on vast, easy to find areas of soft bottom. Furthermore, some heavily studied temperate species are not demersal at all, but are pelagic and live in coastal or oceanic zones (Leis 2006; Leis and McCormick 2002).
The reliance on assumptions and estimations based on temperate studies has been shown to be misleading in a number of studies. Tropical species have a shorter incubation time than temperate species and hatch at smaller sizes and less developed states. Thus they have a longer average pelagic larval duration (Houde and Zastrow 1993; Leis and McCormick 2002). However, tropical species appear to develop more quickly (Job and Bellwood 1996) and possess better swimming performance than would be assumed with temperate species (Leis and Carson-Ewart 1998; Stobutzki and Bellwood 1997). These results may be due in part to the benefits of living in warmer water (Leis 2006).
Further complicating the study of the early life history stages of fishes is the complex, sometimes misleading, and arbitrary terminology present in the literature (Bond 1996; Choat et al. 1993; Leis 1991, 2006). The early life history of fishes is generally divided among egg, larva, and juvenile stages. The larval stage begins with hatching from the egg. During this stage larvae develop into yolk-sac larvae, preflexion larvae, flexion larvae, and postflexion larvae, based on morphological differences (see ‘Biology’ for more in-depth discussion of larval development). What determines the end of the larval stage depends on whether a morphological or ecological definition is used. Utilizing a morphological definition, the larval stage ends with metamorphosis. Metamorphosis is considered to occur once the fins are fully formed, squamations begin and larval characteristics are lost. This process can be abrupt or gradual and may occur independently of the ecologically defined end of the larval stage. The ecological end of the larval stage is called and settlement and occurs when the larva leaves the pelagic realm and moves to a demersal juvenile or adult environment. This review will use a combination of ecological and morphological definitions and refer to larvae as the pre-metamorphic fishes that have yet to settle from the pelagic realm. Thus, post-metamorphosis juveniles which have remained in the pelagic zone (e.g. some carangids and caesionine lutjanids) and the few species that do not have pelagic stages (e.g. Acanthochromis polyacanthus) will not be considered (Bond 1996; Leis 1991).
The acceptance of the “simplifying assumption” has led to larval reef fishes being often termed plankton (Leis 1991, 2006). However, the term plankton carries with it two connotations 1) being of small size, usually <1cm, but sometimes up to a few centimeters and 2) having insignificant swimming and orientation abilities relative to surrounding currents. The larvae of demersal fishes hatch at a small size (a few millimeters) and settle at only a few centimeters, thus fulfilling the size definition of plankton. However, current research has shown that larvae do have the ability to swim against ambient currents and orient themselves to reefs. Although they may be planktonic at the time of hatching, they are clearly nektonic at the time of settlement and some time before. Therefore, the references in the literature to the “planktonic larval stage” and “ichthyoplankton” are both misleading and counterproductive. The neutral term “pelagic” which means resident of the water column more accurately describes the entire presettlement stage of reef fish larvae (Leis 1991, 2006).
Even with generally agreed upon terminology established studying pelagic larval fishes is difficult. Fish larvae occur at very low densities. Even in the protected waters of the Biscayne Bay where larvae are fairly abundant densities only average 1.8 larvae per cubic meter of sea water (Houde and Lovdal 1984). In offshore locations, densities are even lower (Leis 1991). Furthermore, temporal abundance between seasons and even times of the lunar month can complicate sampling programs (Choat et al. 1993; Rooker et al. 1996). Larval fishes are also small (often only a few millimeters in length). But as they grow and develop changes in behavioral capabilities and size differences necessitate that different techniques and tools be used to fully sample the larval ichthyofauna (Choat et al. 1993). Larvae are also difficult to rear and maintain in the laboratory making detailed experimentations challenging (Leis and McCormick 2002).
Once sampling difficulties are overcome and larvae are acquired there is still the problem of the dearth of taxonomic information on larval fishes. The taxonomic knowledge of larvae is very far from complete when compared to the total diversity of fishes (~24,000 teleost species of which about 60% are marine) (Leis 2006). Descriptions of larval development is still ongoing and work that has been done in the past is often incomplete or incorrect (Brogan 1996; Murphy et al. 2007). Species which have been the basis for many scientific studies, even larval studies, have only recently been described. For example, the pomacentrid, the Ambon damselfish, Pomacentrus amboinensis , has been studied in numerous ecological and larval studies, but has only lately been described (Murphy et al. 2007). Species in common and highly conspicuous reef fish families such as Chaetodontidae (Leis 1989), Lutjanidae (Brogan 1996), and Serranidae (Kohno et al. 1993) have been described, but many still lack descriptions.
Despite the obstacles, researchers have learned a great deal about the pelagic stage of larval reef fishes. This review will describe some of the current tools which have been used successfully to study this life history stage of reef fish larvae. After which, the current state of research on the biology, behavior, and ecology of larval reef fishes will be examined. Afterwards, the advantages of a pelagic larval stage will be discussed. Finally, directions for future research will be enumerated.
Bay, L. K., et al. 2006. Intraspecific variation in the pelagic larval duration of tropical reef fishes. Journal of Fish Biology. 68(4): 1206-1214.
Bond, C. E. 1996. Biology of Fishes. Toronto: Thomson Learning. 750pp.
Brogan, M. W. 1996. Larvae of the eastern Pacific snapper Hoplogagrus guntheri (Teleostei: Lutjanidae). Bulletin of Marine Science. 58: 329-43.
Choat, J. H., et al. 1993. A comparison of towed nets, purse seine, and light-aggregation devices for sampling larvae and pelagic juveniles of coral reef fishes. Fishery Bulletin. 91: 195-209.
Hjort, J. 1914. Fluctuations in the great fisheries of northern Europe. Rapports Proces-Verbaux des Reunions, Conseil Permanent International Pour L’exploration de la Mer. 20: 1-13.
Houde, E. D. and C. E. Zastrow. Ecosystem- and taxon-specific dynamic and energetics properties of larval fish assemblages. Bulletin of Marine Science. 53(2): 290-335.
Job, S. D. and D. R. Bellwood. 1996. Visual acuity and feeding in larval Premnas biaculeatus. Journal of Fish Biology. 48: 952-963.
Kohno, H. et al. 1993. Morphological development of larval and juvenile grouper, Epinephelus fuscoguttatus. Japanese Journal of Ichthyology. 40: 307-16.
Leis, J. M. 1989. Larval biology of butterflyfishes (Pisces, Chaetodintidae): What do we really know? Environmental Biology of Fishes. 25: 87-100.
Leis, J. M. 1991. The pelagic stage of reef fishes: The larval biology of coral reef fishes. In: The Ecology of Fishes on Coral Reefs. Ed. P. F. Sale. San Diego: Academic Press. 183-230.
Leis, J. M. 2006. Are Larvae of Demersal Fishes Plankton or Nekton? Advances in Marine Biology. 51: 57-141.
Leis, J. M. and B. M. Carson-Ewart. 1998. Complex behaviour by coral-reef fish larvae in open-water and near-reef pelagic environments. Environmental Biology of Fishes. 53: 259-266.
Leis, J. M. and M. I. McCormick. 2002. The biology, behavior, and ecology of the pelagic, larval stage of coral reef fishes. In: Coral Reef Fishes. Ed. P. F. Sale. San Diego: Academic Press. 171-200.
Murphy, B. F., et al. 2007. Larval development of the Ambon damselfish Pomacentrus amboinensis, with a summary of pomacentrid development. Journal of Fish Biology. 71: 569-84.
Rooker, J. R., et al. 1996. Sampling larval fishes with a nightlight lift-net in tropical inshore waters. Fisheries Research. 26: 1-15.
Stobutzki, I. C. and D. R. Bellwood. 1997. Sustained swimming abilities of the late pelagic stages of coral reef fishes. Journal of Experimental Marine Biology and Ecology. 175:275-86.
