How to observe and report mouth-brooding in cichlid fishes

Those familiar with mouthbrooding cichlids know this behavior is expressed in different ways. A female Satanoperca leucosticta, for instance, lays sticky eggs that after a long interval are picked up by both the male and the female, whereas the female Pseudotropheus zebra alone takes up her eggs just after spawning; and her eggs are loose and do not stick to one another. Some mouthbrooders, such as the Mozambique tilapia, care for the fry for weeks after they have emerged from the mother's mouth, but in marked contrast the fry of Tropheus moorii leave the mother as soon as she releases them. Such diversity indicates an array of evolutionary stories.

In the cedocavus Bujurquina vittata, the male picks up and broods the young for the first five days after which mouth-brooding duties are shared with the female. Photo by Lee Newman.

A recent study (Goodwin et al., 1998) has asked how often mouthbrooding has originated in different lineages of cichlids. To answer this question, the scientists combined cichlid phylogenetic trees that had been constructed by different ichthyologists. Their results suggested that mouthbrooding has evolved independently 10-14 different times. Six to seven of these derivations were identified for lineages of South American cichlids, and from four to seven origins were indicated for African cichlids. These workers' analysis was presented as a starting point, as published observations are rarely systematic and too often lack detail sufficient for conclusive results. So, where can we go from here?

Cichlid observers have an incredible knowledge of the intricacies of cichlid mouthbrooding. Unfortunately, they often fail to appreciate the significance of particularly fine points, i.e., details that are critical to unraveling the various evolutionary paths of cichlid mouthbrooding. To make observers aware of what needs to be recorded, I am proposing a vocabulary for methodically describing mouthbrooding behavior (Table 1). It is virtually a check list of what to record. Note that the categories grade into one another. After all, evolution is a dynamic, on-going process, so some species are bound to be in a transitional state between one category and another.

Modes of Mouth-brooding

No comprehensive vocabulary currently exists for mouthbrooding in cichlids. Some scientists - and many aquarists - use a limited pair of terms: "immediate" and "delayed" mouthbrooding (Keenleyside, 1991). Others write of "primitive" and "advanced" mouthbrooding, but these labels are explanations, not descriptions, and so are to be avoided. To lighten the learning task, I chose Latin roots that, for the most part, are found in everyday English. Nonetheless, for some of the sub-categories, the use of Latin roots seemed cumbersome, so I chose English words.

Cavus refers to mouth; it is Latin for a cavity and is the root of the word cave. I decided not to use the word buccal (L. bucca = cheek), even though one translation of the word is cavity, because I wanted a more general term that includes both the cheek region and the mouth proper. The meanings of postpono-, prompto-, duo- and uni- are obvious. Cedo- has the same root as the English word cede, meaning to give over. Itero denotes repeating a process, as in computer lingo. Semel- conveys doing something only once. For the two sexes, I use matri- for mother and patri- for father. For relationships within a species I use intra- and for between species inter-.

Timing

Postponocavus

Several postponocavus mouthbrooders have now been documented among the monogamous eartheaters, the geophagines of South America (Leibel, 1998), such as the well-known Satanoperca leucosticta. The female lays eggs on a flat surface and after fertilization often covers them with a thin layer of sand (Leibel, 1992). As the moment of hatching approaches (as much as a day or two after spawning in the archetypical example), both parents pull the eggs loose and take the newly-hatched wrigglers into their mouths. Comparable behavior is known for other genera in South America (Mrowka, 1984) and in Africa (Linke and Staeck, 1981).

Geophagus steindachneri from northern South America provides a particularly good example of the kinds of problems in classification that can arise. A female lays adhesive eggs one at a time, but immediately picks each of them up. She departs after the last egg has been laid and fertilized (Leibel, 1993a, b). Using the criterion of adhesive eggs, this species is postponocavus. But applying the temporal criterion, it is promptocavus. Similar difficulties are presented by some African cichlids (Myrberg, 1965; Yanagisawa and Nshombo, 1983).

>In Satanoperca leucosticta, mouth-brooding is postponocavus, i.e., delayed until after the young have hatched. Photo by Lee Newman.

Promptocavus

The prevalent response to freshly spawned, non-adhesive eggs is to pick them up promptly. This is regularly the case among haplochromine and mouthbrooding tilapiine cichlids of Africa. Some promptocavus African cichlids, e.g., Sarotherodon galilaeus, however, may tarry over the clutch for 15-45 minutes, as if to see which parent is going to pick up the eggs (Balshine-Earn and Earn, 1997; Schwanck and Rana, 1991). Various tilapiines in the genus Oreochromis also seem in no hurry to take up the eggs.

Male and Female Investment:

Duocavus

Both parents picking up and retaining the eggs or larvae in their mouths is relatively unusual among cichlids, as though it were an unstable stage in evolution. For instance, among all cichlid species in Lake Tanganyika, only about 15% are duocavus mouthbrooders, whereas in around 45% of the lake's cichlids, the female is the exclusive mouthbrooding parent (Nagoshi and Yanagisawa, 1997). Furthermore, all of the many species of cichlids in Lakes Victoria and Malawi are entirely maternal mouthbrooders (Konings, 1995; Seehausen, 1996).

On the other hand, within the geophagine and other cichlids of South America, mouthbrooding is frequently duocavus (Reid and Atz, 1958; Timms and Keenleyside, 1975; Leibel, 1992). Such species are also characteristically postponocavus mouthbrooders, suggesting the two conditions go together. Duocavus cichlids are found among African riverine species as well (Loiselle, 1985; Schwanck and Rana, 1991), particularly but not exclusively among tilapiine species.

Table I: Classification of Mouthbrooding
Category	Subcategory	Definition
Postponocavus		The parents postpone picking up the adhesive eggs for several hours typically until near hatching
Promptocavus		The non-adhesive eggs are picked up soon after spawning.
Duocavus		Both parents pick up and continue to care for the eggs or fry.
Cedocavus		One parent picks up the eggs soon after spawning, but later cedes them to the other parent.
	1) Male leads	Male is the first parent to pick up the eggs.
	2) Female leads	Female is the first parent to pick up the eggs.
	3) Cede fry	Parent retains young until they can swim; both parents then care for the fry.
	4) Cede wrigglers	Parent gives embryonic young to its mate.
	5) Intraspecific	Parent gives offspring to foster parent of the same species. Also called farming out.
	6) Interspecific	Parent gives offspring to parents of a different species. Also called farming out.
Unicavus		One parent picks up the eggs and becomes their sole caretaker.
	1) Patricavus	The male is the caretaker.
	2) Matricavus	The female is the caretaker.
Iterocavus		After the fry are released, they remain with the parent(s) and re-enter the parent's mouth to escape danger or to overnight.
Semelcavus		After the fry are released, they depart, never to re-enter the parent's mouth.

Cedocavus

Cedocavus species are typically monogamous, and no category presented here illustrates so well the possible progression from biparental to uniparental care, or perhaps the reverse (Kuwamura, 1997). In its simplest expression, one parent picks up the eggs, and it alone carries the offspring for some time while the mate simply hangs around. Then the first parent releases the young in front of the other. Their care from that point on is either the task of the other parent or of both parents.

- Male leads: The sex of the parent that first picks up the eggs differs between widely unrelated mouthbrooders, reinforcing the view that it has been independently evolved. In the South American species Bujurquina vittata, the male goes first and broods the eggs and larvae in his mouth for about five days (Socolof, 1972). Similarly, the male of the West African Chromidotilapia Güntheri (Myrberg, 1965) takes up the eggs and retains the brood until they can swim.

- Female leads: In contrast to the above, the female is the initial mouthbrooding parent in all of the several known cedocavus cichlids found in Lake Tanganyika (Kuwamura, 1986; Kuwamura, 1988; Yanagisawa et al., 1996; Nagoshi and Yanagisawa, 1997; Neat and Balshine-Earn, 1999).

- Cede fry: If the zygotes are sufficiently well developed to swim on their own when first released from the brooding parent's mouth, then both parents regularly guard the fry. When danger threatens, or at nightfall, the fry retreat into the mouths of both parents (Kuwamura, 1988; Myrberg, 1965; Socolof, 1972; Yanagisawa and Nshombo, 1983).

- Cede wrigglers: This behavior is known from a number of monogamous cichlids in Lake Tanganyika. When the wrigglers still possess large yolk sacs at the time the mother-passes them over to the father, he keeps them in his mouth until they are fully developed (Kuwamura et al., 1989; Yanagisawa, 1986; Yanagisawa et al., 1996; Neat and Balshine-Earn, 1999). Usually both parents subsequently guard the fry. In Tanganicodus irsacae, however, parental care ceases with the release of the fry by the father (Kuwamura and Nagoshi, 1987).

- Farming out - intraspecific parents: A special case of cedocavus behavior is "farming out" (Yanagisawa, 1985b). In intraspecific farming out, the parent approaches another parental pair of the same species and spits its fry out into the school of young of the soon-to-be host family. Originally, farming out was thought to be caused by one member of a pair deserting its mate, who is then left alone to raise the young. However, recent observations on Perissodus microlepis have shown that pairs may also farm out their young into the families of other pairs (Ochi et al., 1995).

- Farming out - interspecific parents: Parents in some species farm out their fry to alien species. Ochi (1996) discovered that two of the species he observed in Lake Tanganyika were especially susceptible to receiving foreign fry that were deliberately ceded to them by their alien parents (see also Yanagisawa et al., 1996).

The most remarkable instance of interspecific farming out, however, is the insertion of fry into the brood of a pair of catfish, Bagrus meridionalis, in Lake Malawi. The cichlids here are usually the maternal mouthbrooders Copadichromis pleurostigmoides and Ctenopharynx pictus. The interspecific relationship is complex, including different behavioral adaptations by the cichlids for foisting their fry on the catfish (McKaye et al., 1992). Note that unlike the other examples of cedocavus cichlids, the females in these cichlids are the sole caretaker and are hence unicavus.

The overwhelmingly "mainstream" pattern of mouthbrooding in Rift Lake cichlids is matricavus, as illustrated by Xenotilapia spiloptera Poll & Stewart 1975 from Lake Tanganyika. Photo by Ad Konings.

Unicavus

By far the most common expression of mouthbrooding in cichlids is just one parent brooding the eggs and embryos (Goodwin et al., 1998).

Patricavus: In Sarotherodon melanotheron, the blackchin mouthbrooder, the male alone broods the eggs (Shaw and Aronson, 1954; Aronson, 1949; Oppenheimer and Barlow, 1968). Other species in the genus may do the same (Trewavas, 1983). Recent research has been stimulated by this unusual patricavus behavior in a cichlid, with the possibility that males and females may have reversed their sexual roles (Balshine-Earn and McAndrew, 1995; Barlow, 1991).

Matricavus: Exclusively female care of the brood recurs among cichlids of most lineages of mouthbrooders (Goodwin et al., 1998). It is prevalent, for instance, among tilapiine cichlids of the genus Oreochromis (Lowe, 1956; Trewavas, 1983). The African haplochromines are entirely matricavus (Eccles and Lewis, 1981; Seehausen, 1996). And in the Neotropics, in the few known cases of unicavus behavior, all are matricavus (Newman, 1993; Leibel, 1993b; Leibel, 1993a).

Iterocavus

In many different lineages, a parent - typically, but not always, the mother - continues to accept her young back into her mouth after they have become free-swimming (Baerends and Baerends-van Roon, 1950; Kuwamura, 1986; Yanagisawa, 1985 a). This parent-young relationship can last a number of weeks in some of the tilapiine cichlids (Trewavas, 1983). South American cichlids also have prolonged iterocavus behavior, and they can be either duocavus or unicavus (Leibel, 1992).

Semelcavus

The typical semelcavus mouthbrooder is a single female who lays among the biggest eggs reported in cichlids, carries them in her mouth for up to one month, and releases large, well-developed young that immediately leave the mother when set free, never to return. Lake Tanganyika provides numerous instances of semelcavus parenting (Kuwamura and Nagoshi, 1987; Ochi, 1993; Yanagisawa and Nishida, 1991).

Semelcavus cichlids are also common in Lake Malawi (Ribbink et al., 1980; Balon, 1977). Although semelcavus species are characteristically matricavus, rare exceptions are known (Kuwamura and Nagoshi, 1987)..

Tilapiine cichlids in the genus Sarotherodon have semelcavus representatives, such as the patricavus blackchin mouthbrooder S. melanotheron (Brestowsky, 1968). A related species in this group, Stomatepia pindu from Lake Barombi Mbo, has also been described as semelcavus, but with female care (Loiselle, 1997). These observations hint that most of the genera in the Sarotherodon group could prove to be semelcavus.

Tanganyikan mouth-brooders of the genus Tropheus, such as this T. moorii "moliro" are typically semelcavus, i.e., once the fry are released, they are not taken back into the mouth. Photo by Ad Konings.

Observing and Reporting.

If cichlidophiles were to report on each of the categories in Table I, it would help immensely in working out the various paths that cichlids have taken in the evolution of mouthbrooding. And the reporter should not stop there. She or he should routinely note other bits of information, such as how many eggs were laid (particularly in relation to the size of the female), and how long the eggs were brooded. Ideally, a sample of eggs should be sent to Ron Coleman (contact him at http://www.cichlidresearch.com) so he can add to his egg data base.

References

• Aronson, L. R. 1949. An analysis of reproductive behavior in the mouth breeding cichlid fish, Tilapia macrocephala (Bleeker). Zoologica 34: 133-158.
• Baerends, G. P., and J. M. Baerends-van Roon. 1950. An introduction to the study of the ethology of cichlid fishes. Behav. Supp. 1:1242.
• Balon, E. K. 1977. Early ontogeny of Labeotropheus Ahl, 1927 (Mbuna, Cichlidae, Lake Malawi), with a discussion on advanced protective styles in fish reproduction and development. Env. Biol. Fish. 2:147176.
• Balshine-Earn, S., and D. J. D. Earn. 1997. An evolutionary model of parental care in St. Peter's fish. J. Theor. Biol. 184:423-431.
• Balshine-Earn, S., and B. J. Mc-Andrew. 1995. Sex-role reversal inthe black-chinned tilapia, Sarotherodon melanotheron (Rüppel) (Cichlidae). Behaviour 132:861-874.
• Barlow, G. W. 1991. Mating systems among cichlid fishes, pp. 173190. In: Cichlid Fishes. Behaviour, Ecology and Evolution. M. H. A. Keenleyside (ed.). Chapman and Hall, New York.
• Brestowsky, M. 1968. Vergleichende Untersuchungen zur Elternbindung von Tilapia-Jungfischen (Cichlidae, Pisces). Z. Tierpsychol. 25: 761-828.
• Eccles, D. H., and D. S. C. Lewis. 1981. Midwater spawning in Haplochromis chrysonotus (Boulenger) (Teleostei: Cichlidae) in Lake Malawi. Env. Biol. Fish. 6:201-202.
• Goodwin, N. B., S. BalshineEarn, and J. D. Reynolds. 1998. Evolutionary transitions in parental care in cichlid fish. Proc. Roy. Soc. London (B) 265:2265-2272.
• Keenleyside, M. H. A. 1991. Parental care, pp. 191-208. In: Cichlid Fishes. Behaviour, Ecology and Evolution. M. H. A. Keenley-side (ed.). Chapman and Hall, New York.
• Konings, A. 1995. Malawi Cichlids in their Natural Habitat. Cichlid Press, El Paso, Texas.
• Kuwamura, T. 1986. Parental care and mating systems of cichlid fishes in Lake Tanganyika: a preliminary field survey. J. Ethol. 4:129-146.
• -. 1988. Biparental mouthbrooding and guarding in a Tanganyikan cichlid Haplotaxodon microlepis. Jap. J. Ichthyol. 35:62-68.
• -. 1997. The evolution of parental care and mating systems among Tanganyikan cichlids, p. 5786. In: Fish Communities in Lake Tanganyika. H. Kawanabe, M. Hori, and M. Nagoshi (eds.). Kyoto University Press, Kyoto.
• Kuwamura, T., and M. Nagoshi. 1987. Female-to-male shift of mouthbrooding in a Tanganyikan cichlid fish, Tanganicodus irascae, p. 86. In: Ecological and Limnological Study on Lake Tanganyika and its Adjacent Regions. Vol. IV. H. Kawanabe and M. Nagoshi (eds.). Kyoto University, Kyoto.
• Kuwamura, T., M. Nagoshi, and T. Sato. 1989. Female-to-male shift of mouthbrooding in a cichlid fish, Tanganicodus irsacae, with notes on breeding habits of two related species in Lake Tanganyika. Env. Biol. Fish. 24:187-198.
• Leibel, W. 1992. Goin' south - Part 8: Cichlids of the Americas. Maintaining and spawning the demonfish. Aquar. Fish:27-36.
• -. 1993a. Goin' south - Part 12: Cichlids of the Americas. The key to success with these fish is excellent water quality. Aquar. Fish: 43-51.
• -. 1993b. Goin' south - Part 9: Cichlids of the Americas. Maintaining and spawning the demonfish. Aquar. Fish:45-51.
• -. 1998. The evolution of mouthbrooding in South American geophagine cichlids. Lafayette College, Easton, Pennsylvania.
• Linke, H., and W. Staeck. 1981. Afrikanische Cichliden I - Buntbarsche aus Westafrika. Tetra-Verlag, Melle, Germany.
• Loiselle, P. V. 1985. The Cichlid Aquarium. Tetra-Press, Melle, Germany.
• -. 1997. Aquarium husbandry of the pindu, Stomatepia pindu Trewavas 1972. Cichlid News 6:11-17.
• Lowe (McConnell), R. H. 1956. The breeding behaviour of Tilapia species (Pisces; Cichlidae) in natural waters: Observations on T. karomo and T. variabilis Boulenger. Behaviour 9:140-163.
• McKaye, K. R., D. E. Mughogho, and T. J. Lovullo. 1992. Formation of the selfish school. Env. Biol. Fish. 35:213-218.
• Mrowka, W. 1984. Is the parental-care behaviour of Aequidens paraguayensis (Cichlidae) optimal? Behaviour 89:128-146.
• Myrberg, A. A., Jr. 1965. A descriptive analysis of the behaviour of the African cichlid fish Pelmatochromis Güntheri (Sauvage). Anim. Behav. 13:312-329.
• Nagoshi, M., and Y. Yanagisawa. 1997. Parental care patterns and growth and survival of dependent offspring in cichlids, p. 175-192. In: Fish Communities in Lake Tanganyika. H. Kawanabe, M. Hori, and M. Nagoshi (eds.). Kyoto University Press, Kyoto.
• Neat, F. C., and S. Balshine-Earn. 1999. A field survey of the breeding habits of Eretmodus cyanostictus, a biparental mouthbrooding cichlid in Lake Tanganyika. Env. Biol. Fish. 55:333-338.
• Newman, L. 1993. Maintenance and breeding of the red hump eartheater, Geophagus steindachneri Eigenmann and Hildebrand 1910. Cichlid News 2:14-16.
• Ochi, H. 1993 Maintenance of separate territories for mating and feeding by males of a maternal mouthbrooding cichlid, Gnathochromis pfefferi, in Lake Tanganyika. Jap. J. Ichthyol. 40:173-182.
• Ochi, H., and Y. Yanagisawa. 1996. Interspecific brood-mixing in Tanganyikan cichlids. Env. Biol. Fish. 45:141-191.
• Ochi, H., Y. Yanagisawa, and K. Omori. 1995. Intraspecific broodmixing of the cichlid fish Perissodus microlepis of Lake Tanganyika. Env. Biol. Fish. 43:201-206.
• Oppenheimer, J. R., and G. W. Barlow. 1968. Dynamics of parental behavior in the black-chinned mouth breeder, Tilapia melanotheron (Pisces: Cichlidae). Z. Tierpsychol. 25:889914.
• Reid, M. J., and J. W. Atz. 1958. Oral incubation in the cichlid fish Geophagus jurupari Heckel. Zoologica 43:77-88.
• Ribbink, A. J., A. C. Marsh, B. A. Marsh, and B. J. Sharp. 1980. Parental behaviour and mixed broods among cichlid fish of Lake Malawi. S. Aft. J. Zool. 15:1-6.
• Schwanck, E., and K. Rana. 1991. Male-female parental roles in Sarotherodon galilaeus (Pisces: Cichlidae). Ethology 89:229-243.
• Shaw, E. S., and L. R. Aronson. 1954. Oral incubation in Tilapia macrocephala. Bull. Amer. Mus. Nat. Hist. 103:375-416.
• Timms, A. M., and M. H. A. Keenleyside. 1975. The reproductive behaviour of Aequidens paraguayensis. Z. Tierpsychol. 39:8-23.
• Yanagisawa, Y. 1985a. Mating system of a cichlid fish Xenotilapia sp, p. 22-23. In: Ecological and Limnological Study on Lake Tanganyika and its Adjacent Regions. Vol. 3. H. Kawanabe (ed.). Kyoto University, Kyoto.

This article was originally published in Cichlid News Magazine, Jul-00 pp. 24-31, It is reproduced here with the permission of author George Barlow and Aquatic promotions.