(This article was originally published in Freshwater and Marine Aquarium Magazine, Oct 1982; pp. 14-19, 79-90. It is here reproduced with the permission of author Dr. Paul V. Loiselle).
|A parental male Herotilapia multispinosa of the Los Chiles, Costa Rica population. The contrast rich color pattern serves as an unambiguous warning of unpleasant consequences to potential fry predators. Photo by Paul V. Loiselle.|
(Editor note: The present article, written in 1982, lacks all the recent modifications in Central American cichlid taxonomy. In an attempt to make it current I have updated Dr. Loiselle's article to include all the curently accepted Cichlid genus. I have followed Kullander paper (Kullander 1996) on Heroines with comments for Cichlasomines systematics for this end. May-1997.)
Central America is extremely rich in cichlid species but remarkably poor in cichlid genera. A single representative of the genus Geophagus, G. crassilabrus, occurs as far north as central Panama, while the genus Aequidens, widespread in South America, is represented here only by Ae. caerulopunctatus, which barely manages to squeak into southern Costa Rica. The remaining Mesoamerican cichlids are either representatives of the Heroine group or of genera derived from it. One of these Heroine derivatives, the large piscivore Petenia splendida, I have discussed in a previous article (Loiselle, 1980b). I intend to devote this essay to the remaining cichlid genera endemic to this region, Herotilapia and Neetroplus. Both are attractive cichlids of modest dimensions, whose behavior makes them as interesting to scientists as they are to aquarists.
The genus Herotilapia was established in l903 by the French ichthyologist Pellegrin for a small cichlid described 44 years previously as Heros multispinossus. Albert Gunther, who described this species in 1869 from material collected in Lake Managua, Nicaragua, was apparently more impressed by its elevated dorsal and anal spine counts than by details of its dentition, for he placed it in the genus Heros. Pellegrin chose to place greater significance on the combination of incisor-like anterior and pleuricuspid posterolateral outer jaw teeth in assessing the taxonomic placement of this species. Subsequent workers have followed his lead to the present day.
The generic name Herotilapia suggests a Heroine-like fish with Tilapia characteristics. It is hardly to be credited that Pellegrin, a typical turn-of-the-century museum ichthyologist, could have had a sufficiently detailed knowledge of the natural history of either Tilapia as that genus was then understood or of Herotilapia to have drawn such an analogy directly. His appreciation instead suggests an astute piece of inferential reasoning based upon morphological features shared by the two taxa. In the nearly 80 years since Herotilapia was described, our greater understanding of the biology of both genera has confirmed that these morphological similarities indeed reflect strikingly similar evolutionary responses to ecologically convergent conditions.
Herotilapia is a monotypic genus, comprising the single species H. multispinosa. The genus differs morphologically Archocentrus, whose most familiar representative is the ubiquitous A. nigrofasciatus, only in the nature of its buccal dentition. The elevated spine counts that so impressed Gunther can be regarded as merely the culmination of a trend towards more heavily armored vertical fins well-evident in that subgenus. The sympatrically occurring Archocentrus centrarchus has dorsal and anal spine counts that overlap with those of H. multispinosa, a feature that led to the confusion of the two fish by aquarists in the mid 1960's. Herotilapia and A. centrarchus co-occur from the basin of the Rio San Juan in Nicaragua southward along the Atlantic slope of Costa Rica (Miller, 1966). Unlike A. centrarchus, H. multispinosa is also found on the Pacific slope of Costa Rica, in the basins of the Rio Tempisque and the Rio Bebedero (Bussing, 1976).
The first attempt to introduce H. multispinosa to the hobby was made in 1961 by Dr. William A. Bussing, at that time a graduate student studying the ichthyofauna of Costa Rica. Bill brought several Costa Rican cichlids back with him, including the first Archocentrus septemfasciatum and Hypsophrys nicaraguense to be imported, as well as a strikingly marked race of A. nigrofasciatum from the Atlantic lowlands. All three of the foregoing were successfully bred and distributed by the late Gene Wolfsheimer, but plans to perform the same office for H. multispinosa were frustrated by the death of the single pair's male in transit. The widow spawned regularly until her ultimate demise, giving Gene a chance to photograph the striking parental color pattern of this species. I can still remember my astonishment when Bill showed me a slide of that solitary spawn-tending female. Here was a small cichlid colored unlike any member of the family to date imported. Like everyone else who saw the photo, I came away convinced of the absolute desirability of successfully establishing H. muttispinossa as an aquarium fish.
When Bill returned to Costa Rica, he was able to arrange another shipment of fish to the U.S. In order to assure the survival of some breeding stock, he sent fry the second time around. The shipment was a success, and the fry were distributed to a number of interested parties, among them Guy D. Jordan, a pioneer cichlid keeper and one of the founders of the American Cichlid Association. As Guy's fish grew, their resemblance to the fish photographed by Wolfsheimer diminished. Finally, Guy and Dick Stratton, another of the A.C.A.'s founders, carried out the definitive test of their fish's identity by looking at an individual's jaw teeth with a hand lens. Instead of the diagnostic mixture of incisor-like anterior and pleuricuspid lateral teeth found in Herotilapia, they saw an unbroken series of conical teeth. Taken with their fish's elevated dorsal and anal spine counts, this led them to conclude they actually had a tankful of Archocentrus centrarchus, a diagnosis subsequently confirmed by Dr. George W. Barlow. The initial misdiagnosis by Bussing was almost certainly due to the overlap in these meristic characters that I mentioned earlier. Field identification of fry by sny other means would have been most difficult to accomplish. This circumstance seems to underly the confusion of H. multispinosa and A. centrarchus by American hobbyists in the 1960's. This problem has since been resolved in part by publication of accurately identified photographs of both fish in widely available reference works (Goldstein, 1970, 1973) and in part by the virtual disappearance of A. centrarchus from the ranks of available cichlids.
According to Guy Jordan (pers. comm.), the successful introduction of the rainbow cichlid, as H. multispinosa came to be called, was due not to Bussing's efforts but rather the result of an importation of this species sometime afterwards by a private party in Dallas, Texas. The sporadic export of wild Amphilophus. labiatum from Nicaragua at that time suggests the Rio San Juan basin as the focus of the original aquarium strain of this species. Marie McCann, a correspondent who was aware of his interest in unusual cichlids, sent Guy a pair of these fish in 1964. He had no difficulty in inducing the fish to spawn and distributed fry among his extensive network of contacts. His success was apparently far from unique, for by 1970 H. multispinosa had become so generally available in the United States that it was widely regarded as the quintessential beginner's cichlid!
Due in large measure to the efforts of Dr. Jeffrey R. Baylis, a great deal more is known about the natural history- of the H. multispinosa than is typically the case for a Neotropical cichlid. During his graduate program at Berkeley, Jeff had an opportunity to supplement laboratory work on this species' behavior with field studies. The knowledge he gained about the ecology of the rainbow cichlid allowed him to meaningfully interpret much of the behavioral data derived from his laboratory observations. Though this species has attracted the attention of subsequent workers (Smith-Grayron and Keenleyside, 1978), the results of his research (Baylis, 1974) remain a model of the successful integration of field observation and laboratory studies to which any aspiring student of cichlid behavior can profitably turn for inspiration.
|A sexually quiescent male Archocentrus centrarchus. In the past, this species was confused with H. multispinosa by aquarists. Like the rainbow cichlid, this species has elevated dorsal and anal fin spines counts, but differs dramatically in its coloration and uniformity conical jaw teeth. Photographed at Shedd Aquarium. Photo by Paul V. Loiselle.|
|This parental female Archocentrus centrarchus displays the same sort of reverse countershading that can be seen in comparably motivated Rainbow cichlids. Reverse countershading is characteristic of the breeding dress of all representatives of the genus Archocentrus and of many other Heroine species as well. Photo by Paul V. Loiselle.|
|Algal browsing by this sexually quiescent male Los Chiles Herotilapia multispinissa is facilitated by the distinctive jaw teeth that define the genus. Even outside of periods of sexual activity, the rainbow cichlid is a very attractively colored fish. Photo by Paul V. Loiselle.|
|The rainbow cichlid is not characterized by extreme sexual dimorphism. Females such as this one are best recognized by their plumper appearance and somewhat shorter, more rounded dorsal and anal fins. Among sibs of the same age, females are also typically slightly smaller than their brothers. Photo by Paul V. Loiselle.|
According to Jeff, H. multispinosa is essentially a fish of seasonally inundated marshes and potholes. Juveniles and sexually inactive adults can be found in the shallow, usually vegetated peripheral waters of large lakes and rivers or in the small creeks flowing into them. Sexually active adults and very small fry, however, are to be found only in seasonally flooded habitats. At Los Chiles, Costa Rica, Jeff's main study site, the only other fishes regularly found sharing such habitats with rainbow cichlids were the ubiquitous characin Astyanax mexicanus, a member of the Poecilia mexicana complex of short-finned mollies, and juvenile Parachromis managuense. No other cichlid breeds in such habitats. Indeed, aside from the molly, no other representative of the local ichthyofauna reproduces therein.
The impoverished fish fauna of these flooded areas is a function of the obstacles they pose to successful colonization. As Jeff describes them (Baylis, 1974, 1976), such bodies of water are shallow, extremely turbid and highly eutrophic. They are often found in cow pastures, whose inhabitants regularly contribute to their nutrient load. Those located near villages are not merely eutrophic but downright polluted. His description of one particular collecting site is guaranteed to put any sane individual within earshot off for life of the idea of collecting cichlids in the wild! The great quantities of organic matter in these habitats can lead to severe local deoxygenation, particularly at night, while their extreme shallowness 25.0 cm is about their average depth exposes their inhabitants to daytime temperatures up to 35°C. with a 15°C. nightly drop possible under certain circumstances. Fish living in such shallow water are always at considerable risk from such piscivores as wading birds and water snakes. If this were not enough, a proportion of these pools is ephemeral, disappearing during the dry season with predictable consequences for their finny inhabitants. Clearly, a fish must possess a very special set of characteristics to make a success of colonizing such an environment.
However, there are real advantages to specializing for life under such seemingly unattractive conditions. Precisely because they are shallow and nutrient-rich, primary productivity in such habitats is high. A copious supply of vegetable matter in the form of diatoms, flocculent macrophyte detritus and filamentous algae is available to any fish with the necessary equipment to harvest and process it. The physical characteristics of such a system operate to screen out possible competitors for food, living space and breeding sites. Finally, and possibly most important of all, these pools are totally devoid of predatory fishes large enough to menace an adult Herotilapia while supporting only a restricted spectrum of fry predators. All these factors make such habitats attractive places for a cichlid that barely attains 10.0 cm SL to set up housekeeping.
The rainbow cichlid exploits these difficult but rewarding habitats with a two-phase life history. Reproduction occurs in seasonally flooded habitats during their dry season isolation from their parent bodies. While physical conditions are extreme at this time, their isolation assures the breeding Herotilapia of a greatly reduced complement of fry predators with which to contend. An abundance of food and a dearth of competitors promote rapid growth of their progeny in the bargain. When the rains come, the rising water level of the adjacent rivers brings them into contact with previously isolated pools and marshes. This triggers the dispersal phase of the life cycle, in which the previously impounded rainbow cichlids can move into new habitats. A certain proportion of the fish will enter suitable breeding habitat prior to its isolation by receding waters at the end of the rainy season. These fish will compromise the coming dry season's reproductive population. Those not fortunate enough to colonize such habitats will remain behind in the river shallows, their chance at reproduction lost for another year. Such a life history pattern can be likened to an evolutionary lottery. To succeed, players must disperse into a peripheral habitat suitable for spawning that will persist through the dry season. Losers, defined as fish that either fail to colonize a suitable breeding habitat or else colonize one that proves a fatal trap are penalized a year's reproductive output, or in the extreme case, their lives. The payoff for winners is a disproportionate contribution to the next year class of juveniles.
Such a life history pattern requires of its practitioners a specialized suite of adaptations. The physiological characteristics the rainbow cichlid must possess to prosper in such habitats are obvious enough. Tolerance of high concentrations of dissolved metabolites and indifference to short term fluctuations in temperature and water chemistry head the list. Experiments (Baylis, 1974) have demonstrated this species' ability to tolerate the sort of hypoxic conditions likely to occur in the presence or large-scale decomposition of organic matter. I have already alluded to the dental modifications that allow Herotilapia to collect vegetable foods efficiently. Parallel digestive specialization must accompany such dental features as well. There are no data on the digestive physiology of this species, but the rainbow cichlid does possess the same sort of elongate intestine characteristic of such other herbivorous cichlids as Sarotherodon, Tilapia and many of the mbuna. Withal, the rainbow cichlid as well equipped to survive in seasonally flooded habitats.
However, a successful life history pattern based on the use of peripheral habitats as breeding sites requires that its practitioners be capable of more than mere survival under extreme environmental conditions. They must be able to reproduce successfully under such circumstances as well. The foregoing adaptations are obviously necessary preconditions to any reproductive effort. However, they do not directly address the reproductive process itself. Survival assured, the essential difficulty Herotilapia must surmount is that of breeding successfully under crowded conditions. According to Baylis (l974), space in isolated pools is limited, while rainbow cichlid populations are often dense. To breed successfully under these conditions requires a special set of behavioral adaptations to complement the physiological specializations considered above.
The first such behavioral adaptation is the ability to spawn within a very circumscribed territory. Herotilapia multispinosa can manage comfortably with a territory 30 cm on a side. If its environment is characterized by a diverse topography, pairs can even pull off a successful spawning in a territory 20 cm square (Baylis, 1974). Equally important are mechanisms that serve to attenuate aggressive responses and substitute ritualized interactions for actual physical combat. Quite apart from the risk of serious injury that arise from prolonged, unritualized aggression, such behavior is energetically costly. Spawning under crowded conditions would be virtually impossible under crowded conditions in the absence of some mechanism that modifies the expression of aggressive behavior.
The rainbow cichlid keeps the cost of aggression at a minimum by combining an extremely complex but highly unambiguous visual signaling system with a marked tendency to habituate to the presence of territorial neighbors. The color patterns of Herotilapia allows an individual to communicate its motivational state with great precision. This reduces the likelihood of accidental conflict by minimizing the ambiguity in any exchange of information between two potential combatants. Should physical conflict arise, however, ritualization of the resulting interaction keeps risk of injury minimal. Attacks are directed not at vulnerable targets such as the eyes, but rather towards specific elements in the color pattern located well away from such sensitive areas (Baylis, 1974). Combat between two neighboring males of this species is, to paraphrase Shakespeare, an affair of sound and fury, whose level of significance is quite disproportionate to its relative lack of substance.
Spawn defense is also made easier if the parents can signal their motivational state unambiguously to potential predators. Fry predators that learn to associate the physical abuse they receive from parental cichlids with a specific color pattern are likely to avoid their vicinity in the future. By so doing, they also avoid the vicinity of the pair's mobile fry. Many cichlids practice this sort of aversive conditioning of potential fry predators, but few of them are faced with the problem Herotilapia must overcome in transmitting the necessary signal. The intense golden ochre and black spawning colors that aquarists find so attractive is the rainbow cichlid's solution to the challenge of visual signaling in an extremely turbid environment. Research on underwater optics (Luria and Kinney, 1970; Lythgoe, 1979) has shown that this combination of colors is maximally visible in such environments, and thus best suited for transmitting a signal unambiguously and over a considerable distance.
Finally, to make a success of such a life history pattern, its practitioners must be highly fecund. Entry into a suitable breeding habitat during the dispersal phase is essentially a matter of chance. Thus the more offspring a pair produces per spawning effort, the greater the likelihood that a few of them will penetrate suitable habitats and in turn spawn successfully in the future. The need to channel as much of its available energy to spawning as possible is another reason why selection has favored adaptations that reduce intraspecific in the rainbow cichlid. Herotilapia rises to this challenge by producing the largest spawns of any cichlid in its size range to date studied. Baylis (1974) reports spawns of up to 1500 eggs from a female 10.0 cm SL. In my experience, clutches of up to 2500 eggs are not at all unusual from females in that size range.
The seasonality of reproduction in nature also places a premium on the ability to respawn promptly should anything disrupt a previous spawning effort. According to Baylis (1974), a female rainbow cichlid can respawn from 10 to 14 days after the loss of a clutch of eggs. This reduced recycling time, combined with the extreme fecundity of H. multispinosa has led some killifish and Betta fanciers to reverse the usual polarity of trophic interactions in their fish rooms. Many breeders of these fish like to keep a single large cichlid around as a sort of biological garbage disposal for the numerous culls any program of selective breeding is likely to engender. However, in situations where winter scarcity of live food is a real problem, astute breeders have discovered that a pair of rainbow cichlids constitutes a convenient and highly reliable means of upgrading dried food into an abundant supply of high quality live food for the main objects of their attentions!
It should be fairly obvious from the foregoing why H. multispinosa is such a superb beginner's cichlid. The same abilities that allow this species to prosper in highly eutrophic habitats in nature also allow it to shrug off a good deal of accidental mismanagement in captivity. This is not to say that the basic principles of good cichlid keeping should not be applied to this species. Like other abuse-resistant species, the rainbow cichlid does much better and certainly shows up more satisfactorily when housed under suitable conditions. A temperature range of 21°-25°C. suffices for ordinary maintenance, with a rise to 30°C. advisable for breeding. Dropping the temperature gradually down to 23°C. once the fry are mobile will reduce the likelihood of precocious respawning and its attendant complications, a contingency to which this species is otherwise quite susceptible in captivity (Baylis, 1974; Loiselle, 1982b). So long as obvious extremes in either direction are avoided, Herotilapia is indifferent to pH and hardness. While resistant to mismanagement of the nitrogen cycle, it does appreciate regular partial water changes, to which it responds with enhanced coloration.
Food is also no problem in captivity. Any of the usual live and prepared foods will satisfy the rainbow cichlid's healthy appetite. The intense golden base coloration is at its best when their keeper sees fit to offer his fish a fair amount of plant matter in their diet. Herotilapia will take any of the usual leafy vegetables enthusiastically, provided they are presoftened by a quick dip into boiling water. If one fails to provide them with supplementary plant matter, the fish are capable of making up for this deficiency by nibbling any rooted plants in their aquarium. Duckweed is also likely to come in for such attention, though floating fern (Ceratopteris) is effectively immune. Assuming their appetite for vegetable food is otherwise satisfied, H. multispinosa is no danger to rooted plants. Unlike many Heroine species, the rainbow cichlid confines its digging strictly to periods of reproductive activity.
When not engaged in courtship or spawning, Herotilapia is an excellent community resident. While it will eat fish up to the size of a male guppy, such an eventuality is unlikely to occur in large aquaria. The rainbow cichlid is a most inept predator, lacking both the morphological and behavorial characteristics of a successful piscivore. The response of sexually active Herotilapia towards other fish varies with the species concerned and the size of the breeding tank. Other cichlids and mid-water living fishes are simply interdicted from the pair's territory. In a tank of 100 1 capacity or larger, they run little danger of serious injury. As mentioned earlier, rainbow cichlids can, and typically will, make do with a restricted breeding territory, while in a spacious tank, the other residents have adequate space to move into when the prospective parents begin feeling their oats.
Bottom dwelling species run a considerably greater risk of injury. Herotilapia will systematically persecute individuals of the lurking eleotrid predator Gobiomorus dormitor, attacking individuals even beyond the boundaries of its territory. According to Baylis (1974), such predators will not survive such attentions unless furnished with shelter such as a length of PVC pipe which the cichlids cannot penetrate. Such behavior is understandable enough, given that dormitor is a major source of danger to sympatrically occurring cichlid fry in Central America. However, the average aquarist is usually less than forgiving when this sort of selective obliteration is directed at the catfishes and loaches sharing a tank with a reproductively inclined pair of rainbow cichlids. Heavily armored loricariid catfishes are effectively immune to the consequences of Herotilapia harassment, but prudence dictates removing any other bottom dwellers as soon as a pair of rainbow cichlids begins defending a breeding site.
If only a single pair of rainbow cichlids is present in the breeding tank, the male will assume the role of territorial defender, while the female concentrates on preparation of the spawning site. If a group of Herotilapia is present, the behavior of the pair is more complex and interesting. The male of a pair concentrates on keeping other males out of his bailiwick. This end is achieved through a combination of overt aggression and modification of the territorial borders, themselves, through vigorous excavation. Such environmental modification apparently facilitates territorial defense by creating a sharp topological boundary easily recognized by intruders. In effect, the male is drawing the proverbial "line in the dirt," and woes betide the intruder rash enough to cross it! Females are indifferent to intruding males during the early stages of site defense, but will vigorously repulse intruding females, whose presence is, to the contrary, ignored by the resident male. Indeed, he seems to behave towards such females much as he did towards his own mate during the initial stages of pair formation. This has led Baylis (1974) to suggest that it is chiefly the intolerance of the paired female to intruders of her own sex that precludes evolution of a polygynous mating system in this species.
Both sexes will nip off the potential spawning site, though the female does the bulk of the actual preparation. The pair invariably chooses a sloping or vertical surface. A slate tilted against the side of their tank or an inverted flowerpot is favored alternatives in captivity. Lacking such choices, a pair will usually spawn on the glass wall of their aquarium. This preference in spawning sites reflects another behavioral response to the exigencies of breeding in a soft-bottomed, highly turbid habitat, as it greatly facilitates the task of keeping the developing embryos free of suffocating sediment. The appearance of a conspicuous ovipositor is a reliable predictor of a spawning within 24 hours. The mechanics of spawning follow the usual Heroine pattern (Loiselle, 1980a, l9N2a) and result in the deposition of a large, roughly circular plaque of beige to yellowish-white, ovoid eggs, each measuring c. 2.0 mm along its major axis.
Hygienic behavior towards the spawn is well developed. Both sexes will fan the eggs, though the female is more actively committed to this aspect of their care than is her mate. Though he tends to concentrate on keeping intruders at a distance, the male will swap off roles with the female. Such behavior in captivity is particularly evident at feeding time. It allows both parents to maintain a normal food intake during a period when such activity would be virtually impossible in nature. This may be one reason why rainbow cichlids are prone to respawn in the presence of an earlier brood when bred under aquarium conditions. The fish go through a very predictable "greeting ceremony" (Baylis, 1974) whenever one member of the pair returns to the territory after an absence of more than a few seconds. This behavior is depicted in the accompanying series of photos.
The sequence begins with the arrival of the male and simultaneous departure of the female, the smaller of the two fish, and ends with her as the exclusive custodian of the spawn. The particular interaction took place over an interval of three minutes. Note the vertical placement of the large egg plaque. Photos by Paul V. Loiselle..
At 29°C. the fry are chewed out of their surrounding eggshells 24 hours posthatching. Both parents participate in this phase of the reproductive cycle and share in the transport of the wrigglers to a nursery area prepared in advance. This is usually a pre-dug pit, located some distance from the spawning site. If the bottom is too soft to allow the construction of such a pit, or else too heavily overlain with detritus, the parents will "plaster" the wrigglers against a vertical surface (Baylis, l974) or else hang them in the roots of floating aquatic plants such as Ceratopteris. While it may seem quite uncichlid-like, such behavior is not unique to H. multispinosa. The sympatrically occurring Archocentrus centrarchus will also place its wrigglers in fine leafed plants in captivity (Schwarz, 1977), as will the Brazilian species 'Cichlasoma' oblongum (R. Vanner, pers. comm.). Comparable behavior has been recorded in the wild for Mesonauta festivum (Lowe-McConnell, 1969). Such a choice of nursery area is also typical of angelfish and discus when they are allowed to rear their young without human intervention (Innes, 1948; Lindaman, 1935; Sterba, 1966; Wagner, 1958; Winter, 1934). The adaptive significance of such "hanging" or "plastering" behavior is unclear for the other species from which it has been reported, but it is a safe bet that in the rainbow cichlid, it serves to keep the fry in the best-oxygenated portion of an often severely hypoxic habitat.
The wrigglers may be moved several times during the five to six day interval between hatching and the attainment of mobility. The parents seem more apt to move their progeny in circumstances that have led them to eschew placing the wrigglers in a pre-dug pit. The presence of other fish also seems to elicit frequent shifts of the fry from one location to another. The two sexes are equally active in this aspect of parental behavior. They also share the task of fanning the wrigglers. Close coordination between the male becomes even more evident once the young are fully mobile and begin to swarm out of their former shelter. Aggression directed towards other fish is particularly intense just prior to this point. It may have disagreeable consequences if the breeding tank is too small to allow potential victims to move beyond the spatial limits set on its expression.
The mobile fry are easily fed. They are large enough to take Artemia nauplii, microworms and finely powdered prepared foods for their first meal. The pair will also stir up accumulations of detritus for their foraging young. I have also seen the adults masticate chunks of food and spit the resulting cloud of particles into the swarming mass of young. As is the case in many Heroines, the parents exert very little control over the movements of their offspring. The only signal that elicits a consistent response from the fry is a motor pattern, Baylis (1974) refers to as "jolting." Here the displaying individual jerks its head sideways and snaps its vertical fins erect while holding its fully outspread pectoral fins at a right angle to the body. The ventral fins are then slowly folded against the body. The entire performance requires less than a minute and may be repeated several times in rapid succession. The fry respond by swimming towards the signaling parent and dropping to the bottom. This response attenuates rapidly as the fry grow older. By the end of their first week of mobile existence, it is effectively extinct. By this point in their ontogeny, the fry in captivity cease to school coherently and instead forage in a diffuse swarm. Such behavior certainly complicates their guardians' task.
The adults clearly suffer from an inability, arising from such behavior by the mobile fry, to provide a complete defense against predators during this phase of the reproductive cycle. I suggest that the extreme aggressiveness pairs' display towards other fishes, beginning at a point well prior to the actual spawning act, has evolved in part as compensatory response to this state of affairs. As mentioned earlier, the contrast-rich color pattern of sexually active rainbow cichlids facilitates the process by which potential threats to the fry come to associate a sound thumping with a specific visual stimulus. The victims of such aversive conditioning are unlikely to seek out the vicinity of parental Herotilapia thereafter. If begun early enough, such behavior reduces the probability that the foraging fry will encounter predators as they move about under their parents' surveillance.
That the ideal result of completely eliminating this source of mortality goes unattained is evinced by the steady attrition of the swarm of fry when rainbow cichlids spawn in a community situation in captivity. No less than by the simple observation that the inhabitants of Nicaragua and Costa Rica are not up to their armpits in Herotilapia! I have no idea what percentage of a successful spawning survives to independence in nature. In captivity, however, given a tank of 200 1 capacity or larger, it is not unusual for ten to fifteen percent of a brood to attain this point under community conditions. Given the fecundity of the rainbow cichlid, this is a most impressive performance.
Baylis (1974) reports another facet of parental behavior likely to enhance fry survival. He found that parental Herotilapia are quite tolerant of the presence of other pairs' fry among their own, provided the newcomers were about the same age as their own offspring. Indeed, in one instance, the most aggressive of three concurrently breeding pairs actively kidnapped the entire reproductive output of the remaining two. Within a week of the fry having attained mobility, this pair was defending a single enormous school of small fish. This may at first glance seem an instance of parental behavior run amok. A little reflection, however, will reveal that from an evolutionary perspective, there is a method to such madness.
Predation upon defended cichlid fry is essentially a random process, determined by the probability of encounter between a predator and any given fry. Thus the best way to reduce mortality among fry is for their parents to behave in a manner that reduces the probability of such encounters. Aversive conditioning of potential predators, as mentioned earlier, is one means of accomplishing this end. To the degree that it is successful, it will keep predators well away from the mobile young. Active defense of the school of fry should ideally deter any predators who failed to get the original message. However, no defense is perfect, and penetration by a few predators daily is inevitable.
This being the case, the next available tactic for minimizing the risk of a pair's fry providing a predator's next meal is to dilute them in someone else's. As the number of alien fry in such an amalgamated swarm increases, the probability that as predator will take one of a pair's own offspring decreases. Thus, up to the point where the enlarged swarm of fry becomes inordinately burdensome for two fish to supervise or becomes itself an overly enticing lure to predators, a policy of, "In our little school, there's room for one more," is a most advantageous one for a pair to pursue. However, as older, larger young are themselves potential predators on younger fry, there is nothing to be gained from tolerating their presence. Hence the size-selective aspects of such adoptions alluded to earlier. Baylis (1974) was the first to suggest such adaptive significance to this behavior. His insight was most convincingly vindicated by McKaye and McKaye (1977) who found that both intra- and interspecific adoption of alien young was a common occurrence among the cichlids of Lake Jiloa in Nicaragua.
The duration of fry defense in nature in unknown. Parental care can persist up to eight weeks in captivity. The actual interval is usually shorter than this, due to the tendency of pairs to respawn within three to four weeks of a previous effort. I have explored the probable causes of this phenomenon in an earlier essay (Loiselle, 1982b). It is impossible to predict how a pair will behave towards their older fry under such circumstances. Thus, it is wisest to separate parents and offspring as soon as the telltale signs of impending spawning manifest themselves. Given plenty of food and frequent partial water changes, the fry grow very rapidly. Herotilapia multispinosa can begin breeding at sixteen weeks posthatching. It thus enjoys the somewhat dubious distinction of being the most sexually precocious Neotropical cichlid! The advantages of such precocity to an animal that breeds in unpredictable and often ephemeral environments are obvious. The advantages to an aquarist with limited tank space of working with anima1s 3.5 cm SL should be equally so!
In 1970, a color mutant of the rainbow cichlid characterized by a novel distribution of black pigment on the body and fins was first discovered. This new color form made its appearance in an aquarium shop in Waterloo, Iowa, among a population of tank-reared Herotilapia being held for sale. This heteromelanic color variety is generally known as the Norton black-trim rainbow cichlid, in honor of Dr. Joanne Norton, who worked out the genetics of this phenotype and was largely responsible for distributing it among interested cichlid keepers. According to Norton (1975), the mutant gene behaves as a simple Mendelian recessive. It would be interesting to know if the black trim phenotype exists in nature. Given the degree to which the normal color pattern of this species is adapted to its optical environment, one would predict with a fair degree of confidence that the mutant form would be at a considerable disadvantage in all pre-reproductive and reproductive interactions relative to the normal form. The black trim rainbow is as easily maintained in captivity as the wild, or as Norton (1975) refers to it, the "drab" form. From the perspective of one who has kept both, I find myself in disagreement with Norton's assessment of the relative attractiveness of the two fish. I find the normal form a much more colorful cichlid. Fortunately, in such matters one can be guided by one's personal taste. Trahit sua quemque voluptas!
Since its aquaristic debut twenty-one years ago, Herotilapia multispinosa has progressed from a much sought-after rarity to a respected member of that group of old stand-bys referred to by experienced cichlidophiles with a mixture of affection and condescension as "beginner's cichlids." A species must possess a rare combination of features to merit such a distinction, and it goes without saying that not every new arrival qualifies for membership in this rather exclusive fraternity. I would never, for example, predict such status for Neetroplus nematopus. The bases of this prediction will be elaborated in the concluding segment of this essay.
- Baylis, J.R. 1974a. The behavior and ecology of Herotilapia multispinosa (Teleostei. Cichlidae). Z. Tierpsychol. 34: 115-146.
- Baylis, J.R. 1976. A rainbow in muddy water. Tropical Fish World 3(l): 8-14.
- Bussing, W. 1976. Geographic distribution of the San Juan ichthyofauna, with remarks on its origin and ecology. In T.B. Thorson (Ed.) Investigations of the Ichthyofauna of Nicaraguan Lakes. School of Life Sciences, U. of Nebraska, Lincoln, pp 157-175.
- Goldstein, R.J. 1970. Cichlids. T.F.H. Publications. Jersey City.
- Goldstein, R.J. 1973. Cichlids of the World. T.F.H. Publications, Neptune City.
- Gunther, 1869. An account of the fishes of the States of Central America based on collections made by Capt. J.M. Dow, F. Godman, Esq. and O. Salvin, Esq. Trans. Zool. Soc. Lond., 6(7); 377-394.
- Innes, W. 1948. Exotic Aquarium Fishes. Ninth edition. Innes Publishing Co., Philadelphia.
- Kullander, Sven 1996, "Heroina isonycterina, a new genus and species of cichlid fish from Western Amazonia, with comments on cichlasomine systematics." Ichthyological Explorations of Freshwaters, Vol 7, No. 2, pp 149-172, 13 figs., 5 tabs.
- Lindman, H.R. 1935. Symphysodon discus. Recording their first successful spawning as accomlished by Gustav G. Armbruster. Fish Culturist (Philadelphia) 14(10): 227-230.
- Loiselle, V. 1980a. Cichlasoma septemfasciatum A new color form of a Central American Cichlid. FAMA 3(1): 44-49 et seq.
- Loiselle, P.V. 1980b. Giant predatory cichlids: The true guapotes. FAMA 3(8): 39-47 er seq.
- Loiselle, P.V. 1982a. Our national cichlid Cichlasoma cyanoqunatum (Baird and Girard 1854). FAMA 5(5): 6-11 et seq.
- Loiselle, P.V. 1982b. Cichlid Potpouri.FAMA 5(9): et seq.
- Lowe-McConnell, R.H. 1969.The cichlid fishes of Guyana, South America, with notes on their ecology and breeding behavior. Zool. J. Linn. Soc. 48: 255-302.
- Luria, S.M. and J.A. Kinney. 1970. Underwater vision. Science l67: 1454-1461.
- Lythgoe, J.N. 1979. The Ecology of Vision. Oxford University Press, Oxford. McKaye, K.R, and N.M.
- McKaye. 1977. Communal care and kidnapping of young by parental cichlids. Evolution 31; 674-681.
- Norton, 1975. The black-trim rainbow cichlid. Buntbarsche Bull. (51): 10-12.
- Pellegrin, 1903. Contribution a 1'etude biologique et taxonomique des poissons de la famille des Cichlides. Mem. Soc. Zool. France (8)/6: 41-402.
- Schwarz, 1977. Archocentrus centrarchus (Gill and Bransford 1877).Amer. Cichlid Assoc. Cichlid Index 2(9): 1-2.
- Smith-Grayton, P.K. and M.H.A. Keenleyside. 1978. Male-female parental roles in Herotilapia multispinosa (Pisces: Cichlidae). Anim. behav. 26: 520-526.
- Sterba, 1966. Freshwater Fishes of the World. Revised English language edition. Studio Vista Ltd., London. Wagner, O. 1958. Behavior of discus fish in spawning. The Aquarium (Philadelphia) 27(5): 136-143.
- Winter, 1934. Breeding Symphysodon discus. The Aquarium (Philadelphia) 3(4); 89-90.