Featured Article

Introduction

Even the largest coral reef (Fig. 1) and the biggest coral colony start life as a diminutive pelagic larva, and the choices that such larvae make with regards to where they settle (Fig. 2) have consequences that cascade through the entire reef ecosystem. A coral reef clearly is more than the sum of the component corals, but without a clear understanding of the biological events affecting the early life stages of coral it is impossible to fully understand the events that maintain coral communities. This article describes what is currently known about these events, and collectively considers coral larvae, recruits and juvenile colonies as early life stages.

Are all small corals young?

Early life stages of corals are, by definition, young relative to adult colonies, and are also small in size. It is important to note however, that the reverse is not always true, that small corals are not necessarily young. This paradox arises from the prodigious capacity of corals to reproduce asexually through partial mortality and fragmentation (Fig. 3). Partial mortality occurs when, for example, a disease kills a portion of a coral colony leaving behind patches of tissue separated by dead skeleton. Fragmentation, in contrast, generally occurs through physical breakage of colonies by waves and storms, and this can also create small pieces of coral scattered across the reef. In both cases (partial mortality and fragmentation), the processes create small colonies that are not chronologically old, and importantly, they share the host genotype of the parent colony. Thus, small colonies are not “early life stages” per se, and of greater biological significance, they do not contribute to genetic diversity of the population. These negative effects of asexual proliferation can be accentuated by impaired sexual reproduction, which can be caused by colonies transitioning from sexual maturity to sexual immaturity as they become smaller.

The distinction between early life stage and small pieces of older colonies has considerable ecological relevance, because disturbances can have strong and positive effects on the population density of small pieces of older colonies. At the same time, such disturbances reduce the capacity to increase the number of early life stages through sexual reproduction. For instance, a severe storm can populate a reef with vast numbers of fragments of living corals (i.e., small corals), all of which are the same age as the parent colony from which they broke, and some of which (usually the biggest) will survive the breakage event. A protracted period of chronic disturbance can also generate large numbers of smaller colonies through the more insidious effects of partial mortality. Under such conditions, grand old coral colonies (that also can be very large) can progressively break into a large number of small pieces (Fig. 3), which subsequently have a high chance of dying (as described below).

More than a quarter century ago this unusual relationship between size and age in reef corals was described in a well-known article that questioned whether corals lie about their age. There is no doubt that corals can indeed, lie about their age, and therefore any analysis of early life stages of corals must establish that such colonies are genuinely young.

The source of early life stages of corals

All early life stages of corals start life as larvae (Fig. 4), but the larvae are not all created equal. Apart from the differences in larvae expected based solely on species-level differences, the larvae fall into two groups depending on whether they originate from parents practicing brooding or broadcasting reproductive patterns. In brooding corals, the eggs are retained in the maternal polyp. Although most corals are hermaphrodites, sperm originating in a different colony typically fertilizes the eggs within the polyps of brooding corals. Thereafter, the larvae are brooded until they are relatively well developed, and only then are they released to the water. Such larvae usually receive their symbiotic algae from the parent (maternal or “vertical” inheritance), small numbers of lipid-rich larvae are produced, and the released larvae generally are able to settle to a suitable surface almost immediately upon release. Thus, for brooded corals it is possible for their larvae to settle close to the parents, and indeed aggregated patterns of small colonies are often a clear sign of this reproductive strategy. Importantly however, while brooded larvae are competent to settle almost immediately following releases, they have the capacity to remain in the water column for lengthy periods, with the results from one species suggesting this time can exceed 100 days. In virtually all cases however, it is still uncertain how long brooded larvae remain free swimming in the natural environment.

In broadcasting corals, both the eggs and sperm and released into the water column where typically they rise rapidly to the surface. The positive buoyancy of lipid-rich eggs aids this motion, and often eggs are packaged with sperm in “egg-sperm” bundles, so that both gametes rise simultaneously to the surface of the ocean. Once there, the eggs and sperm from multiple colonies mix with together, with the process aided by the breakdown of egg-sperm bundles wherever these are formed. Fertilization of the eggs leads to rapid development, and within about 72 hours a free-swimming larvae is formed. This obligatory period of development means that larvae of broadcast spawning corals are not immediately competent to settle, and instead require a lengthy period before they are ready to do so. During this period, the larvae mature and initiate a symbiotic relationship with algae taken up from the water column (i.e., “horizontal” inheritance).

There are further striking differences between brooding and broadcasting corals, not least of which is that they differ in the biogeographic regions where each strategy is most prevalent. Brooding corals are particularly common in the West Atlantic and Caribbean, whereas broadcast spawning corals are most common throughout the Indo-Pacific. It is not clear what has lead to this striking difference over evolutionary time, but there is a strong likelihood that it is related to the historic instability of the Caribbean basin compared to the larger Pacific. Unlike the Pacific, on a historical time scale of tens of thousands of years, the Caribbean has experienced extended periods of cool and turbid conditions that generally are unfavorable to reef corals. Corals with brooding life history strategies appear to have faired better than broadcasting corals during such adverse conditions, perhaps because their larvae are retained within the polyp during the most sensitive early developmental stage. If this interpretation is correct, it might be important to evaluating the response of coral reefs to global climate change (GCC). Importantly, this biogeographic trend in distribution of life history strategies might suggest that brooding corals will survive better than broadcasting corals as the negative effects of GCC intensify.

A further dramatic difference between the two modes of reproduction is that broadcast spawning is associated with tight synchronization across entire regions, such that large areas of reef spawn on the same night, or perhaps over several nights. This phenomenon is best illustrated on the Great Barrier Reef of Australia, where corals along the entire length of this massive reef spawn on the 3^rd through 6^th nights following the full moon in late spring (usually in November). This event is so vast that it has been described as the “greatest sex show on earth”, and it generates such large quantities of gametes that they can be seen from airplanes as brightly colored surface slicks. If these gamete slicks are trapped within enclosed bays, they can kill fish and other marine organisms through the development of anoxia as the gametes decay. Similar synchronized mass spawning occur throughout the world, and even in the Caribbean, the relatively poorly represented broadcast spawners tend to release their gametes on a few nights at the end of August.

Although there are a few corals that reproduce all year, the majority of species, regardless of reproductive strategy, reproduce in tight synchrony with a variety of environmental signals. Seawater temperature is an important factor dictating when corals reproduce, with many corals spawning gametes or releasing larvae as the seawater warms in the spring, or even when it is reaches its greatest annual temperature. Lunar synchronization also is important for reproduction in many species, with gamete spawning or larval release associated with specific phases of the moon. Such tight synchrony presumably is achieved through detection of moonlight (amazingly, that they are able to detect!) or tidal stages. In all cases of reproductive synchronization, the proximal causes appear to be a combination of factors including temperature, photoperiod, and lunar phase, but the ultimate cause (i.e., the evolutionary cause) presumably is enhanced reproductive success. In part, enhanced reproductive success is almost certainly one result of gamete concentrations increased through synchronized spawning, but it is also likely to be affected by the coincidence with abundance pelagic food supplies or perhaps the availability of algal symbionts in the seawater. The avoidance of predictably adverse conditions such as heavy rainfall or temperature anomalies may also provide selective advantages for exquisitely timed reproduction. For some species, synchronized spawning can be so tightly controlled that it can separate gametes of closely related species, even if spawning times differ by only a few hours on a single night. Thus, in some cases the ultimate purpose of reproductive timing might be to maintain barriers to hybridization among species.

Coral larvae and coral recruits

If coral larvae are to be successful at forming new coral colonies, sooner or later they must settle out of the water column and initiate a benthic existence as a sessile polyp. This might occur rapidly in the case of brooded larvae, or after a period of development in the case of larvae originating from spawned gametes. In both cases, settlement can potentially be delayed by weeks or months, presumably if conditions are not suitable, or might be accelerated by warmer seawater. Accelerated development has important consequences, as it shortens larval longevity, and therefore also affects dispersal distances. Already there is evidence that rising seawater temperature associated with GCC will lead to important changes in the development, mortality, and dispersal of pelagic larvae like those of reef corals.

When coral larvae are ready to settle, they typically engage in complex behaviors that help to ensure that they settle in suitable locations. The selective value of such behaviors is likely to be very strong, as most corals have but one chance to find a suitable location. Ultimately, it is the settlement of larvae that determines the environmental conditions that the coral colony will be exposed to for its entire life. Although it is uncertain where coral larvae spend their time within the water column, specifically how long they spend in surface layers where light levels are high and damaging UV radiation is intense, at some point they descend to the reef surface. This behavior is a result of the combined effects of several tactic responses including avoidance of bright light, preference for certain colors of light (such as the blue wavelengths found at depth), and gravity.

Once in close proximity to the seafloor, coral larvae have exquisite preferences for different types of surfaces, with the choice presumably made to increase the chances of survival. Typically, larvae avoid unstable surfaces like sand and flexible algae, and instead favor hard surfaces such as rock, dead corals, and coralline algae. There is strong evidence that the larvae of many corals prefer only a few species of crustose coralline algae (conspicuous to humans as pink and hard crusts [Fig. 5]), of which Titanoderma spp. serves as a particularly strong inducer of larval settlement. Such algae have been described as serving as “chemical fly papers” for coral settlement, and they contain an alcohol-soluble chemical that is strongly favored by settling larvae. Should such biological inducers of coral settlement become less common on coral reefs, as has been suggested in some locations, the consequences could be dire for future reef growth. Recently however, several researchers have presented evidence that the attractive properties of coralline algae for coral larvae may also be a result of bacteria closely associated with the surface of the algae.

When a suitable surface is found by coral larvae, they initiate the settlement and metamorphosis processes whereby the cigar-shaped larvae changes into a compressed ball that quickly resembles a coral polyp having a diameter of about ½ mm. It is hard to overstate the complexity of these processes, as they involve a fundamental reorganization of the body structure from a pelagic larvae to a benthic polyp, the production of feeding tentacles and, significantly, the initiation of calcification. In some species, the first elements of the calcareous skeleton are formed within a few hours of settlement, and quickly the outline of the coral cup and the first septae are visible. Presumably calcification continues as the soft tissues grow and begin to feed autotrophically (by photosynthesis) and heterotrophically (on plankton and bacteria), and the symbiotic algae start to accelerate the rates of calcification as they do in adult colonies.

One enigmatic aspect of coral settlement is that numerous studies have shown that settling larvae prefer cryptic habitats (versus open and conspicuous locations). In shallow water, this is manifest as settlement on undersurfaces where competition with algae is avoided, but in deeper water (where algae grow less vigorously) this trend is weakened, or disappears. Typically, these trends have been found using settlement tiles – which typically are unglazed terracotta tiles – secured horizontally to the reef surface, with the result that corals settle almost exclusively on the lower surface (Fig. 6). While a good reason for this pattern is that it promotes the survival of coral recruits by removing them from competition with algae, it also creates a two-fold dilemma. First, cryptic settlement removes the small corals from light that is required for photosynthesis and calcification, and second, it places them in a location where their slow growth will make it challenging to “reach” the open reef. Perhaps the answer to this dilemma lies in the possibility that coral recruits are more dependent on bacteria and dissolved food than we currently believe, or that settlement tiles do not provide an ecologically relevant indication of where corals settle on natural surfaces.

The earliest stages of benthic corals (i.e., the recruits) suffer high mortality (sometimes approaching 100% in the first year), but if they survive, slowly they add new polyps and develop into a colony. This process is accompanied by further secretion of limestone and gradually the small colony adopts of the species-specific morphology of the adults.

Coral recruits and juvenile corals

True coral recruits (i.e., corals that appear in the coral population and are detectable by human observers) are notoriously difficult to count because of their small size, cryptic location, and low densities. Some of these difficulties can be overcome by completing surveys using UV light or other specialized lighting that causes coral recruits to fluoresce brightly as white spots against a darkened background. These techniques have promise in the future to measure coral recruitment, but in their absence, many researchers have chosen to work with a larger size class of corals that usually are described as “juveniles”. The definition of such corals varies among studies, but usually they are limited by an upper size limit of about 4-5 cm diameter (Fig. 7).

Juvenile corals are under strong selective pressure to grow fast, because they are exposed to high risks of mortality when small. Unlike larger adult corals that have the “potential” for partial mortality, juvenile typically are either killed completely when exposed to severe disturbances, or they survive (they are too small to experience partial mortality). Little is known about the relative importance of different sources of mortality in killing juvenile corals, but presumably most are killed by grazing by fish and sea urchins, overgrowth by algae, smothering by sediments, physiological stress arising from high temperature and light, or disease.

The high risks of mortality when small creates the potential for an “escape in size”, and therefore juvenile corals are thought to allocate a large portion of their resources to rapid growth. In this context, “rapid” might be in the order of 10 mm/y, although juveniles of branching corals grow considerably faster than those of massive morphologies. There is some evidence – at least from the Caribbean – suggesting that juvenile corals now are growing more slowly than they did 30 years ago, perhaps with as much as a 50% reduction having taken place over this period (Fig. 8). It is unknown what has caused this trend, or whether this trend is consistent throughout tropical seas, but it does raise the disturbing prospect that the rising temperatures and elevated concentrations of carbon dioxide (CO₂) may already be affecting the performance of juvenile corals.

Presumably the strong selection for rapid growth in juvenile corals results in the allocation of substantial carbon and energy resources to this process. Interestingly, this seems to be accomplished by drawing on the resources vested in tissue reserves, and as juvenile corals increase in size (up to 4 cm diameter), there is evidence from the Caribbean that their tissues become thinner and “less fat”. This trend appears to be similar to the seasonal trend for changes in coral biomass, regardless of coral age/size, where coral tissues are fat and thick in the late winter, and become thin and “lean” by the late summer.

Seawater temperature appears to play a strong role in the biology of juvenile corals, with the effects presumably mediated through photosynthesis, calcification, and perhaps diseases. Ecologically, long-term surveys from the Caribbean suggest that the population density of juvenile corals increases in years that (on average) are relatively warm, but these corals subsequently exhibit reduced growth and elevated rates of mortality. Thus, in years that are exceptionally warm, such as 1998 in the Caribbean, the growth rates of juvenile corals can be reduced (or even cease), with the densities of juvenile colonies depressed for some years afterwards. Clearly, the proposed effects of temperature on juvenile corals do not act in simple ways, and the outcomes can be counterintuitive. Further analyses of the effects of temperature on the growth of juvenile corals suggests that the effects may be more profound that simply altering rates of growth. In the Caribbean for example, it seems that temperature alters the ways in which growth changes with colony size in small corals. In years when seawater is cooler than average, there is a strong growth advantage to becoming larger (bigger corals grow disproportionately faster than smaller corals), but this effect disappears in warmer years. Manipulative experiments suggest that this effect is caused by the influence of temperature on the thickness of the tissue layers.

Further Reading

• Bak, R.P.M., Engel, M.S., 1979. Distribution, abundance and survival of juvenile hermatypic corals (Scleractinia) and the importance of life history strategies in the parent community. Marine Biology 54: 341-352.
• Edmunds, P.J., 2000. Patterns in the distribution of juvenile corals and coral reef community structure in St. John, US Virgin Islands. Marine Ecology Progress Series 202: 113-124.
• Edmunds, P.J., 2004. Juvenile coral population dynamics track rising seawater temperature on a Caribbean reef. Marine Ecology Progress Series 269: 111-119.
• Edmunds, P.J., 2006. Temperature-mediated transitions between isometry and allometry in a colonial, modular invertebarte. Proc R Spc B 273: 2275-2281.
• Edmunds, P.J., 2007. Evidence for a decadal-scale decline in the growth rates of juvenile scleractinian corals. Marine Ecology Progress Series 341: 1-13.
• Harrison, P.L., Wallace, C.C., 1990. Reproduction, dispersal and recruitment of scleractinian corals. In: Dubinsky Z (ed) Ecosystems of the world, Vol 25. Coral Reefs. Elsevier, New York, p 133-207.
• Highsmith, R.C., 1982. Reproduction by fragmentation in corals. Marine Ecology Progress Series 7: 207-226.
• Knowlton, N., 2001. The future of coral reefs.Proceedings of the National Academy of Sciences, USA 98: 5419-5425.
• Richmond, R.H., 1997. Reproduction and recruitment in corals: critical links in the persistence of reefs. In: Birkeland C (ed) Life and death of coral reefs. Chapman & Hall, New York p 175-197.