The species of the genus Cynotilapia differ in one main anatomical characteristic from those of Pseudotropheus, Metriaclima,and Tropheops: they have widely spaced unicuspid (conical) teeth in the outer jaws while the other genera have bicuspid (two-pointed) teeth. The scientific name of this genus means “dog-tilapia” and alludes to the possession of these conical, canine teeth.

    Cynotilapia sp-mbamba-LuwinoReef, Mw-male-AKonings
    Cynotilapia sp. Mbamba – Rafa Luwino, Malawi, samiec

    Cynotilapia are frequently encountered in the deeper regions of the sediment-free rocky habitat. Males are territorial and defend small caves as spawning sites. Several species prefer very dark, large caves. The peculiarity of Cynotilapia is that they feed predominantly on phytoplankton. Most mbuna feed on plankton when available but Cynotilapia, especially the non-territorial individuals, feed predominantly in the water column. Males, however, remain much closer to the substrate and sometimes feed from the biocover as well—but never in an efficient manner because the large teeth are not good at harvesting algal strands.

    Cynotilapia sp-mbamba-MbenjiIsland, Mw-female-AKonings
    Cynotilapia sp. Mbamba – Wyspa Mbenji, Malawi, samica

    Like the majority of the mbuna, Cynotilapia spawn in seclusion. The male has to combine his preference for plankton (feeding in mid-water) with the need for a suitably dark spawning site. Such requirements are met in the caves found among large boulders. Females and non-territorial males form large schools in the open water and feed on plankton. At dusk, when the plankton migrates to shallower water, Cynotilapia hide among the rocks, resuming their position in the water column at dawn.

    Cynotilapia sp-mbamba-NkhataBay, Mw-AKonings
    Cynotilapia sp. Mbamba – Zatoka Nkhata, Malawi

    The widely spaced, conical teeth of Cynotilapia are thought to be a relic of another feeding behavior (Fryer & Iles, 1972) and are thus not a specialized feature. Sharp, conical teeth appear to be present in large specimens of M. zebra (Trewavas, 1935) but I have never been able to find them in preserved specimens of any Metriaclima. Maybe the large Cynotilapia that dwells in deep water (40-120 m) and which is sometimes caught together with deep-dwelling utaka, has been confused with M. zebra. Plankton can, in fact, be collected even without teeth, but males may need sharp teeth as persuading agents in their territorial defense! So, unicuspid teeth could equally have developed at a later stage in the evolution of these species.

    Cynotilapia sp-mbamba-MbenjiIsland, Mw-male-AKonings
    Cynotilapia sp. Mbamba – Wyspa Mbenji, Malawi, samiec

    If the feeding technique of Cynotilapia is compared with that of Metriaclima then it becomes apparent that the former is rather unspecialized when compared to other herbivorous mbuna. Cynotilapia sucks in plankton (mostly drifting diatoms, but also zooplankton) and does not need a specialized feeding apparatus or tooth structure to do so. When Cynotilapia feeds from the aufwuchs, as it sometimes does, it picks in a way similar to the pluckers of Pseudotropheus, never in the combing manner of Metriaclima.

    Cynotilapia sp-chinyankwazi-ChinyamweziIsland, Mw-male-AKonings
    Cynotilapia sp. Chinyankwazi – Wyspa Chinyamwezi, Malawi, samiec

    There are several characteristics seen in Cynotilapia that I regard as primitive. Cynotilapia are commonly found in very deep water and are often one of the few species present on reefs deeper than 50 meters. It is often a characteristic of an old species that it is found all around the lake and/or at much deeper levels than those where most of the competition for food and space takes place. The species that have evolved special features to exploit a particular biotope often out-compete those that haven’t, within the biotope in question. The trade-off is that the primitive species is more versatile and can survive, if not thrive, in more and less advantageous places.

    Cynotilapia sp-mbamba-MbenjiIsland, Mw-male-AKonings
    Cynotilapia afra – MaraRocks, Malawi, samiec

    Because Cynotilapia are found in open water and sometimes at relatively deep levels, it could be argued that these species are more easily able to spread to remote areas than other mbuna. However, this argument does not seem to hold for Cynotilapia sp. ‘chinyankwazi’ which is found around the two islets, Chinyankwazi and Chinyamwezi, in the southeastern arm of the lake, and also at a deep reef between Chinyamwezi and the mainland. These habitats are geographically isolated by deep (>100 meters) water, not only from the mainland but also from each other. Although several species of Cynotilapia exhibit dramatic geographical variation and even polychromatism (yellow or white dorsal fins) within some populations, the three populations of C. sp. ‘chinyankwazi’ are indistinguishable from one other. C. sp. ‘chinyankwazi’ probably populated these three habitats before they became geographically isolated from the mainland (i.e. when the lake level was much lower). At the moment these habitats are completely isolated, preventing a steady influx of individuals from neighboring populations. The populations at all three habitats are very large, guaranteeing a large gene pool and thus stabilizing the characteristics of this species.

    Cynotilapia sp-mbamba-MbenjiIsland, Mw-female-AKonings
    Cynotilapia afra – MaraRocks, Malawi, samica

    Two species of Cynotilapia, C. afra and C. sp. ‘mbamba’, are often found sympatrically, but because of the enormous geographical variability of both species it is sometimes difficult to identify them with confidence. Although these species have different habitat preferences—C. sp. ‘mbamba’ occurs in more sediment-rich biotopes—they are often found side by side (syntopic).

    Cynotilapia sp-chinyankwazi-ChinyamweziIsland, Mw-male-AKonings
    Cynotilapia afra – Mala Point, Mozambik, samiec

    C. afra has a very wide distribution along both sides of the lake. For a long time it was the only described species in this genus, and it also has the widest distribution. It is found at Likoma and Chizumulu and further north along the entire northwestern coast to Ngara, and on the eastern coast from Makanjila Point to Chuanga, at Tumbi Point, and from Lumbaulo to Ikombe. The populations at Jalo Reef and Mbenji Island, previously referred to as C. sp. ‘yellow dorsal’ or C. sp. ‘jalo’, are also now believed to belong to C. afra. The area in between Lumbaulo and Tumbi Point harbors at least three forms of Cynotilapia but at present these are difficult to classify. For instance C. afra at Mala Point is very elongate and resembles the population at Pombo Rocks almost 200 km further north, and that at Mara Rocks on the other side of the lake. In a neighboring population at Cobwé, 10 km north of Mala Point, C. afra has a very attractive but completely different color pattern—this variant has been exported under the trade name of “Afra Edwardi” or just “Edwardi” And at Mbweca, about 15 km south of Mala Point, two different species of Cynotilapia are found. One is C. afra, with a more “standard” color pattern, and the other is probably a variant of C. sp. ‘mbamba’. Between Wikihi (Mozambique) and Undu Reef (Tanzania) I have found a very small Cynotilapia that occurs in the intermediate and sediment-rich habitats. At Hai Reef and Chiwindi this form has a very attractive orange dorsal. These populations are assigned to C. afra for the time being. So, not only in coloration but also in the shape of the body there exists geographical variation.

    Cynotilapia sp. Chinyankwazi – Wyspa Chinyamwezi, Malawi – uzębienie

    The distribution pattern of Cynotilapia sp. ‘mbamba’ is similar to that of C. afra but is far from continuous. Along the northwestern coast it is sympatric with C. afra at most locations. In Tanzania, however, it has been seen only at Undu and Lundu, and in Mozambique south of Metangula as far as Malopa in Malawi. The populations in the southeastern part of the lake previously referred to C. sp. ‘black eastern’ are now regarded as belonging to C. sp. ‘mbamba’ as well. A species at the Mbenji islands, previously thought to be another member of Cynotilapia (C. sp. ‘black dorsal’), now appears to be a Metriaclima. However, C. sp. ‘mbamba’ is present at one of the islands (Penga Penga) at Mbenji.

    Cynotilapia afra-Chiwindi, Moz-male-AKonings
    Cynotilapia afra – Chiwindi, Mozambik, samiec

    C. afra has settled in virtually every rocky or intermediate habitat within its range, with males defending small caves among medium sized rocks. It has even established itself in sediment-rich habitats such as those at Hai Reef, Chiwindi, and Cobwé. The enormous variability in male coloration is remarkable, not only within the species as a whole but also within individual populations. Blue-barred males with yellow, orange, white, or black dorsal fins are found only meters apart. Female coloration, however, does not seem to vary much throughout the entire distribution. Unlike those of M. zebra, females lack the vertical bar pattern almost completely and have a light blue-brown color overall. Females of C. sp. ‘mbamba’ have a silvery body, and in those populations where males have yellow dorsal fins the females also have yellowish dorsals.

    Cynotilapia afra-Cobwé, Moz-male-AKonings
    Cynotilapia afra – Cobwe, Mozambik, samiec

    The reason that male coloration is so variable in some populations could be because of male-male competition. If a yellow-dorsal male can’t recognize a black-dorsal one as a conspecific competitor, and vice versa, you can pack twice as many territorial males in the same area. The yellow-dorsal males will only, or mainly, fight with other yellow-dorsal males and ignore the black-dorsal ones. It could even be possible that females have a preference (or are learning to get a preference) for a particularly colored male. And thus, C. afra may actually be in the process of sympatric speciation. I believe, however, that it is not yet that far evolved, at least not in comparison with some species of Cyprichromis in Lake Tanganyika where males either have a yellow or a blue tail, but no individuals with a mixture of either of these two colors. In the case of C. afra there are males with intermediate coloration between the two extremes in each population. So we can continue calling them C. afra. Enjoy your cichlids!

    Cynotilapia afra-JaloReef, Mw-male-AKonings
    Cynotilapia afra – Rafa Jalo, Malawi, samiec


    Fryer, G. & T.D. Iles (1972) The cichlid fishes of the great lakes of Africa. TFH Publications, Neptune, New Jersey.

    Konings, A. (2007) Malawi cichlids in their natural habitat, 4th edition; Cichlid Press, El Paso, Texas.

    Trewavas, E. (1935) A synopsis of the cichlid fishes of Lake Nyasa. Ann. & Mag. N. Hist. 10(16): 65-118.