Banggai cardinalfish


Osnivača sajta
Staff member
9 Avgust 2003
New Plymouth New Zealand
Notes on the biology, geographic distribution, and conservation status of the Banggai cardinalfish Pterapogon kauderni Koumans 1933, with comments on captive breeding techniques.

The aim of this article is to bring to the hobbyist and aquarium community a summary of what is known to date about different aspects of the biology and conservation status of Pterapogon kauderni. In addition, I hope this review will help readers to appreciate its uniqueness among coral reef fishes, and will contribute to raising awareness with respect to the threatened status of its populations and habitat.


The information presented here represents some of the results obtained during the last several years at the New Jersey State Aquarium (NJSA) in our research on the Banggai cardinalfish. The laboratory phase began in early 1997 and focuses on its reproductive biology, ecology, biogeography, evolution, and conservation, and included the development of a captive-breeding program. The fieldwork, which began in 2001 and continued in 2002, has been carried out in collaboration with Mark Erdmann (NRM/EPIQ, Sulawesi, Indonesia) and with financial support from the NJSA, The American Zoo and Aquarium Association (AZA), and The National Geographic Society.
For a detailed description on reproductive biology (including mating behavior, embryology, and juvenile development) see Vagelli (1999), for ecology and conservation see Allen (2000), Erdmann & Vagelli (2001), and Vagelli & Erdmann (2002).


The first time science learned of P. kauderni was in February 1920 when Walter Kaudern, during his 1917-1920 expedition to the Celebes Island (Sulawesi), collected two specimens close to the shore of Banggai Island and sent them to the Leiden Natural History Museum. Thirteen years later Frederick Koumans described the specimens as a new species and genus. He created the genus Pterapogon on the basis of the higher number of rays of the second dorsal fin, which distinguishes this genus from the rest of the apogonid genera.
In his 1972 work on Apogonidae, T. H. Fraser created the subgenus Pterapogon to differentiate the Banggai cardinalfish from Quinca mirifica, a cardinalfish from Northwestern Australia described by Mees in 1966, which Fraser classified within the genus Pterapogon (and subgenus Quinca). In November 1994 ichthyologist Gerald Allen and underwater photographer Roger Steen traveled to Banggai Island hoping to find this beautiful species, which fish guidebook author Kal Mullet had found in that area about two years earlier. In 1995 Allen and Steen published their observations on some ecological aspects of this species, and were the first to speculate about this species lacking a planktonic dispersal phase, and its connection with a very restricted geographical distribution. In 1997 the NJSA began a comprehensive research project on this species.
In 2000 a small population was detected in the Lembeh strait (North Sulawesi), the probable result of an accidental escape from holding nets that an exporter kept in that location.

Reproductive Biology

P. kauderni has a unique set of reproductive characteristics that makes it an exception among coral reef teleosts, e.g., it lacks a planktonic interval (its eggs do not enter into the plankton, and it lacks a larval period); therefore, this species does not have planktonic dispersal; it has very low fecundity; the male mouthbroods the egg-clutch for about 20 days, and after hatching, the embryos remain in the parent oral cavity for about 10 days until completion of development and release as juveniles. The brooding male does not eat during the entire incubating period, which lasts about 30 days.


P. kauderni is a mouthbrooder in which transient pairs are formed and only the males incubate the eggs. Mating is preceded by an elaborate courtship that lasts normally several hours, but may extend to a few days. Prespawning behavior is composed of several courtship displays (female trembling, warping, side to side swimming, nuzzling and mouth-opening by the male).
The most characteristic prespawning display is the side-by-side trembling by the female (see Vagelli, 1999 for a detailed description of pre- and post-spawning behaviors). Side-by-side trembling normally lasts several hours. The only sign of a male's receptiveness is a conspicuous darkness along the lower jaw and sporadic mouth opening displays.

Spawning and Egg Transfer

P. kauderni reproduces all year long under laboratory conditions. Pair formation and courtship usually begin early in the morning and mating occurs during the first hours after noon. There is no secondary sexual dimorphism in this species. Therefore, the only method to differentiate sexes is to keep several individuals together and wait for the formation of pairs. Eventually, mature females will display an enlarged abdomen. Also, during pre-spawning behavior sexes are easy to differentiate because of their distinct courtship displays.
Once the pair is sexed, each fish can be distinguished utilizing the number and position of the silver-colored dots on their abdomens which have a specific pattern in each individual).
Signs of imminent spawning are an increase in the frequency of female approaches during side-by-side trembling, occasional trrembling and opening of the male's mouth, and the pair's proximity to the bottom. Egg release occurs when both individuals are situated side by side, separated about 1 to 2 cm and a few centimeters from the bottom. Occasionally, at the moment of the egg release, the male is located a few centimeters ahead of the female. When about 3/4 of the egg mass protrudes from the female's urogenital papilla (which takes between 1 and 1 1/2 seconds) the male immediately turns around and gulps the clutch by pulling the eggs from the female. The extraction of the eggs by the male requires some amount of pulling, and some eggs are cut off from the dutch and remain attached to the female for a few seconds until they fall off. Egg transfer is completed in no more than two seconds.
If the pair is kept with other specimens during the mating process, the mating female chooses and separates a receptive male :rom the rest of the school, and the pair establishes a home site that is aggressively defended. However, often a secondary male (SNMis allowed to intervene in the courtship. Unlike other approaching individuals, which are immediately chased, the SM is allowed to approach them. The SM follows the pair with a submissive posture, moves away from the pair, and maintains its distance when the pair swims in his direction. Then the SM participates in defense of the site, alone or with one member of the pair. The SM displays trembling behavior (rarely observed in the primary male, PM) and mouth opening to the female. If the PM responds to the female's displays (by remaining close to her and swimming to the bottom), the PM mates and takes the eggs while the SM does not intervene. However, if the PM does not show interest in the female's displays by swimming away and remaining at the middle-top section of the tank for extended periods), the female starts periodically leaving the PM and displaying mating behavior to the SM. The SM responds by swimming to the bottom, and frequently trembling. If the PM does not intervene, then the female occasionally mates with the SM.
Considering the loss of some eggs during the egg-transfer and the normal mortality during embryo development, the expected fertility rate is about 60% (if the breeding pair is kept isolated under captive breeding conditions). If the brooding male is kept with other specimens during the incubation period, then he is usually under heavy stress due to harassment by the other individuals and by the stimulation at feeding time, with a consequent significant decrease of the fertility rate (generally, as spitting of eggs). Since the average number of mature ova before ovulation is about 60 oocytes, about 40 is the average number of juveniles released by an adult male. The highest observed number of released juveniles by one male in our lab was 62, and the highest frequency of deposition by a female observed was once every 25 to 30 days. Latest observations based on field specimens show that the reproductive cycle of P. kauderni has a lunar periodicity. Mating takes place during a few days preceding, during, and few days after full moon (preliminary data indicates that it also has a small reproductive peak duringthe new moon). The release of juveniles from the brooding male has the same periodicity.


Ovulated oocytes measure between 2.7 and 3.0 mm in diameter. An average clutch consists of about 45 eggs that form a round mass 1.5 cm in diameter. The eggs are held together by strong filaments that originate on each egg's capsule and extend outward, entangling with filaments from other eggs. The eggs are ovoid and at spawning are bright yellow-orange with a yolk full of small lipid globules.

Embryos and Juveniles

Embryos hatch after 19 to 20 days of incubation at a postflexion state, and at 5.0 to 6.0 mm SL. Newly hatched eleutheroembryos (free embryos) have a large bilobed yolk and remain for a maximum of 10 days in the male's oral cavity, feeding endogenously until released. The release of most juveniles occurs between the 6th and 10th day after hatching, however, sometimes eleutheroembryos are expelled during the first few days. These premature juveniles are not capable of swimming and generally show signs of slow development. At release, juveniles are of 8 mm SL and their yolk is almost absorbed. Immediately after release, juveniles swim together forming a small compact school and are capable of eating items as large as Artemia nauplii. Observations from wild specimens agree with our lab studies in showing females maturing at about 35-36 mm SL, which corresponds to an approximately age of nine months.

Juvenile Requirement of a HUFA-enriched Diet for Normal Development

Studies in our lab have shown that during the first 3 to 4 months after release, juveniles are very susceptible to a deficit of highly unsaturated fatty acids (HUFA) in their diet. The lack of adequate HUFA intake produces a typical shock syndrome, characterized by sudden, extremely rapid, short, and jerky bursts of motion, brief spiral swims, and falling to the bottom with very expanded opercules and a considerable increase in ventilation frequency. Sometimes, juveniles slowly recuperate after a period of about 5 to 10 minutes. However, many suffer a very violent shock and die just a few seconds after falling to the bottom. This shock syndrome can be triggered by a moderate increase in stress, e.g., feeding, bumping on the tank, turning on the lights, etc. A HUFA-enriched diet reduces significantly the shock events, and decreases mortality to almost zero.
A simple method to deliver HUFA and a very efficient way to avoid the shock syndrome is to feed juveniles for the first three months with enriched newly hatched brine shrimp (Artemia sp.). During the first two months we feed juveniles with enriched Artemia (after hatching Artemia is enriched for 24 hours prior to feeding). After the second month, juveniles are fed a mix of enriched Artemia and very finely minced krill (Euphasia pacijica, E. superba) and mussels (Perna canaliculus). Older juveniles are fed adult Artemia (enriched with microalgae and HUFA supplement) and finely chopped krill and mussels.

Geographic Distribution and Ecology

Our work on geographic distribution included a survey of the entire Banggai Archipelago, the Sula Spur region, and in the area of central Sulawesi from Botok to Luk. In total we searched 86 sites localized in 37 Islands and two reefs. P. kauderni has a highly endemic natural distribution. The entire species is restricted to less than 10,000 km2. We found it off the following 16 islands: Banggai, Bakakang, Bangkulu, Bangko, Bokan, Labobo, Loisa, Masepe, Melilis, Peleng, Tempau, Limbo, Masoni, Mangoa, Seku, and Taliabu. Also, a small population inhabits Luwuk harbor in central Sulawesi. This population is restricted to a very small enclosure, the harbor itself, with very poor environmental conditions (besides daily spills of boat fuel, significant freshwater runoff, human waste, and all sorts of garbage are being dumped into the harbor).
In addition, a small (but growing), recently introduced population was found in the Lembeh Strait (North Sulawesi), approximately 400 km northwest of the Banggai Archipelago.

Habitat Preference

P. kauderni inhabits mainly protected bays on shallow reefs and seagrass beds. However, it also inhabits environments with strong surge and moderate currents. Its preferred depth range is between 0.5 and 4.5 m, but it is most commonly found between 1.5 and 2.5 m with water temperature ranging from 28° to 31°C (82° to 88°F). We found populations in very clear waters in coral reef areas associated with branching corals, while others inhabiting very still murky areas with very poor visibility. In several sites we found them in sandy patches associates with anemones, and on a few occasions in open habitats (low branching corals and rubble). We encountered about 60% of the groups living on coral reefs, 28% on sea grass beds, and about 12% on open areas of sand and or rubble.


Osnivača sajta
Staff member
9 Avgust 2003
New Plymouth New Zealand

P. kauderni is a site-attached fish that remains very close to various hiding substrates, which are very close to the bottom. Some common microhabitats include sea urchins (Diadema setosum, Tripneustes sp.). soft corals (Nephthea sp.), anemones (Actinoden.dron sp., Entacmaea quadricolor, Heteractis crispa, Macrodactyla doreensis, Stichodactyla haddoni), hard corals (Acropora sp., Anacropora sp., Porites sp., Goniopora sp., etc.), hydrozoans (Millepora sp.), and mangrove roots (Rhizophora sp.). Our census data shows that about 49% of individuals were associated with hard corals, about 34% with sea urchins, and 17% with anemones. Although adults and juveniles do not differ in their microhabitat election (i.e., 35.7% of adults and 34.2% of juveniles were associated with urchins, 49.9% of adults and 48.9% of juveniles were associated with hard corals, and 14.7% of adults and 16.8% of juveniles were associated with anemones), newly released juveniles have very distinctive association preferences: 62.5% were found sheltering within anemones, 25% among long branching corals, and 12.5% among the long spines of Diadema urchins.
When sheltering in anemones and Heliojungia corals, P. kauderni often associates with various species of anemone fishes (Amphiprion) and anemone shrimps (Peridimenes), and with several species of cardinalfishes of genera Apogon, Cheilodipterus, and Sphaeramia when living among spines of sea urchins.
In seagrass beds (predominantly composed of the seagrass Enhalus acoroides), P. kauderni is associated with anemones, isolated live corals, sea urchins, or sea stars, forming small fish communities, commonly with Abudefduf sexfasciatus and several species of Pomacentrus, Chromis, and Halichoeres.

Population Structure

During our fieldwork we observed several thousand individuals and carried out eight censuses. During the census work we counted a total of 1273 individuals in 195 groups. The five largest groups encountered in the census areas consisted of: 83 (75 juveniles, 8 adults), 47 (all juveniles), 38 (all juveniles), 31 (28 juveniles, 3 adults), and 29 (all juveniles) individuals. The average group size was about 6.5 individuals.
We observed large groups in only two of the 28 sites where we found P. kauderni, and they were not in the census areas. One of the groups (450 individuals) inhabited around a large Actinodendron anemone in a dense Enhalus acoroides seagrass bed in very calm, low-visibility waters. The other large group, which was associated with branching Acropora corals in clear waters and strong surge, was divided into three subgroups of 150, 200, and 150 individuals. Medium size groups of 25 to 80 individuals were commonly associated with sea urchin clumps, anemones, and corals. Smaller groups of up to 15 individuals were most common, including numerous groups of 3 or fewer individuals.
Most of the individuals censused (79%) were large juveniles between about six and nine months old, 20% were adults, and less than 2% were newly released juveniles less than about 20 mm TL. Newly released young were associated with sea urchins Diadema setosum, corals Hdiofungia actiniformis, and low branching Acropora. One of them was observed staying very close to a one-centimeter-long anemonefish Amphiprion ocellaris in a Stichodactyla haddoni anemone that also harbored adult anemonefish. Newly released individuals were often observed deeply sheltered within substrates, and were never observed in close proximity to conspecific adults or large juveniles. On only one occasion did we encounter fry less than 15 him TL (two weeks or less after release), and they were sheltered among tentacles of a Hdiofungia coral.

Feeding Habits

In contrast to the rest of apogonids that are classic examples of nocturnal feeders among coral reef fishes, P. kauderni feeds mostly during daytime. It is a generalist planktivore-carnivore and, although it preys mainly on planktonic copepods, its diet included many benthic organisms (about 10% of the total copepod intake is represented by benthic copepods). It seems that P. kauderni exhibits some degree of density-dependent food exploitation, and the results of stomach content analysis show that at least in some occasions it behaves as an opportunistic forager. The natural diet of P. kauderni (as indicated by the results of the quantitative analysis) consists of the following major components: copepods = 81.5% (harpacticoids 31.6%, calanoids 29%, cyclopoids 18.1%, siphonostomatoids 0.87%, unidentified 1.91%); isopods = 4.8% (epicarids: 2.72%, gnathiids 1.73%, anthurids 0.14%, unidentified 0.12%, flabelliferans 0.08%); cirripedians = 2.4%; ostracods = 2%; urochordates = 2%; eucarids = 1.5%; mollusks = 1.2% (gastropods 0.97%, bivalves 0.24%); and amphipods = 1.1%. Other taxa less commonly present in its diet are teleost larvae, chaetognathans, polychaetes, acarids, tanaidaceans, cumaceans, mysids and chironomid larvae. The size range of food items varies from about 0.1 mm (calanoid copepods) to 14 mm ,'teleosts, megalop larvae).


We found 27% of individuals (from 28 sites) carried endoparasites in their digestive system and general body cavity. Adult forms of trematodes were found in 29 individuals (9.3%) from 13 locations, and 21 individuals (6.8%) from 13 locations had nematodes. Also, 22 individuals (7%) from 12 sites carried an encysted form of isopod attached to both the external stomach wall and to the internal body wall, and 15 individuals from 7 sites carried a (so far unidentified) type of worm. No ectoparasites were found.


Collecting Pressure

Buyers pay the equivalent of one to 2V2 U.S. cents per fish to local collectors. In turn, buyers are paid about 10 to 12 cents per fish by exporters, most of whom are located in either North Sulawesi or Bali. The high mortality associated with fish transport is the justification (given by buyers) for the ten-fold difference between the price paid by the buyers to the fishers and their selling price to the exporters.
In Bangkulu Island, six local fish-buyers are purchasing approximately 10,000 individual Banggai cardinalfish per buyer per month. Independent estimates obtained from exporters in North Sulawesi confirmed that between 50,000 and 60,000 individual P. kaudemi are received for export each month in North Sulawesi alone. We observed large holding nets throughout the archipelago, some of them containing several thousand Banggai cardinalfish waiting for the buyers. Our collecting data agrees with the findings of Lunn & Moreau (2002) who, after their trade survey in Sulawesi, Bangkulu, and Banggai in March-April 2001, concluded that a minimum of 700,000 fish were traded in the Banggai region in 2000-2001.

Population Size Estimates and Population Status

Based on our census data and the maximum potentially habitable area, the size of the present P. kauderni population (species) is estimated to be about 1.7 million individuals. Given our consistent density estimates of approximately 0.03 individuals per square meter and the small geographic range of P. kauderni, the export estimates of at least 600,000 to 700,000 fish per year are indeed a cause for serious concern. Furthermore, according to Lunn & Moreau (2002), the number of people involved in the trade of P. kauderni has been increasing since 1998, and local fishers reported having more difficulties in finding P. kauderni now than when they first began collecting them.
Several aspects of its biology make this species especially susceptible to indiscriminate collecting, e.g., its attachment to benthic substrates greatly facilitates its capture, thus local fishers collect hundreds of individuals without any special equipment or great effort. Because of the lack of planktonic dispersal and the particular geomorphology of the area (deep channels/trenches and strong currents separate most of the islands), it is almost impossible for this species to expand its geographic range, or to re-colonize areas where they may soon become extinct. Despite long searches specifically directed to find newly released recruits we only detected very few groups of very few juveniles each (generally no more than four to six). Since the number of juveniles released by males in captivity is typically about 40, these observations suggest that P. kauderni likely suffers a high mortality (predation?) during the first few weeks after release. Finally, collecting during the weeks around full (and probably new) moon will have an especially damaging effect since it is the time at which most of the spawns (and juvenile release) take place.
P. kauderni is a rare example of a marine fish with an extremely limited geographic range. This characteristic, coupled with its low fecundity and heavy collecting pressure for the aquarium industry, had lead this species to a dire conservation prognosis. Very few, if any, other marine species are currently subjected to this combination of pressures. We know that P. kauderni populations are being depleted throughout its small range. Without urgent protecting measures, P. kauderni could face extinction in its natural habitat just a few years after its rediscovery by Western science.

Protecting measures in situ are not easy to implement. Restricting the demand from the importing countries is extremely important in trade-related conservation actions. It can be done by responsible self-imposed conservation attitudes from the aquarium industry, or by importation bans. There is no need to remove more specimens from the wild. Captive breeding for this species is a clear option available to any aquarium and hobbyist.
We are currently planning a biodiversity study and a conservation project in association with the Philadelphia Academy of Natural Sciences and the Environment Sympathizer Foundation (Luwuk, Indonesia) aimed to protect the Banggai cardinalfish and its habitats, but to organize such a project and to obtain funding takes time, time that unfortunately this species might not have.

Literature Cited

Allen, G. & R. C. Steene. 1995. Notes on the ecology and behavior of the Indonesian cardinalfish (Apogonidae) Pterapogon kaudemi Koumans. Revue fr. Aquariol. 22: 7-9.

Allen, G. 2000. Threatened fishes of the world: Pterapogon kauderm Koumans, 1933 (Apogonidae). Env. Biol. Fish. 57: 142.

Erdmann, M. & A. Vagelli. 2001. Banggai cardinalfish invade Lembeh Strait. Coral Reefs. 20: 252-253.

Fraser, T. H. 1972. Comparative osteology of the shallow water cardinal fishes (Perciformes: Apogonidae) with reference to the systematics and evolution of the family. Ichthy. Bull. 34: 1-105.

Koumans, E 1933. On a new genus and species of Apogonidae. Zool Med. Mus. Leiden. XVI: 78-79.

Lunn, K & M. Moreau. 2002. Conservation of Banggai cardinalfish populations in Sulawesi, Indonesia: an integrated research and
education project. Final Report. Zoological Society of London (Unpublished report).

Vagelli, A. 1999. The reproductive biology and early ontogeny of the mouthbrooding Banggai cardinalfish, Pterapogon kauderm
(Perciformes, Apogonidae). Env. Biol. Fish. 56: 9-92.

Vagelli, A. & M. Erdmann. 2002. First comprehensive ecological survey of the Banggai cardinalfish, Pterapogon kaudemi. Env. Biol. Fish. 63:1-8.

Alejandro A. Vagelli, a zoologist from La Plata University, Argentina, has managed the research and conservation program at the New Jersey State Aquarium for the last 10 years. He has done extensive studies on the reproductive biology and ecology of diverse marine iff organisms including gobies (Gobiosominni), cottids (Hemitripteridae), and cnidarians (Scyphozoa), and has directed a captive-breeding program in which more than 40 species of marine fishes and invertebrates were reproduced. He has conducted fieldwork in Patagonia, the Ecuadorian Amazon, the Florida Keys, and lately in Indonesia, and is finishing his Ph.D. in environmental science at Drexel University,