Taxonomy & naming
George Albert Boulenger described the fish in 1906 as Paratilapia frontosa, working from a single specimen (BMNH 1906.9.8.140) netted at Kigoma, Tanzania, on one of the early Tanganyika expeditions. It was later transferred to Cyphotilapia, the combination Eschmeyer's Catalog of Fishes treats as valid. The genus name joins Greek kyphos, "hump," to Tilapia — itself a Latinization of the Tswana tlhapi, "fish" — while the epithet frontosa marks the prominent forehead; older synonyms a reader may still meet include Paratilapia frontosa straeleni and Pelmatochromis frontosus.
For most of the twentieth century the genus was monotypic, with frontosa its only member. Takahashi and Nakaya broke that in 2003 by describing Cyphotilapia gibberosa from the southern half of the lake, restricting true Cyphotilapia frontosa to the north. A follow-up study (Takahashi, Ngatunga & Snoeks 2007) resolved the northern six- and seven-banded morphs into one species: the seven-banded form is confined to Kigoma on the eastern shore, and the six-banded form holds the rest of the northern lake. The cleanest way to tell frontosa from gibberosa is the scale count between the upper and lower lateral lines — two rows in frontosa versus three in gibberosa — backed up by frontosa's slightly more elongate, less deep body. Because the trade sells nearly everything as plain "frontosa," famous southern variants like Blue Zaire, Mpimbwe and Moba are taxonomically gibberosa even when the label says otherwise. Before that 2003 split, aquarium literature from the 1990s routinely described southern populations under the frontosa name too — a naming lag worth remembering when reading older sources, since collection points now attributed to gibberosa (the Zambian populations among them) were then still labeled "frontosa."
Morphology
This is one of the largest rock-dwelling cichlids in Tanganyika: a tall, laterally compressed body crossed by broad dark bars over a pale-to-steely-blue ground, with blue often washing over the head and fins and, in northern fish, a little yellow in the dorsal. The trademark is the nuchal hump, a fatty bulge on the forehead that grows with age and peaks in dominant males — females build a smaller version too, so the hump reads as a badge of dominance and maturity rather than a clean sex marker.
Reported lengths depend on how they were taken. FishBase gives a maximum of about 13 in (13 in) total length, while Seriously Fish lists a maximum standard length of roughly 8.7–10 in (8.5–10 in), which leaves out the tail. In a tank, well-kept males commonly pass 12 in (12 in) and females stay smaller at around 8–11 in (8–11 in); males also grow longer, with more extended dorsal and anal fins. The field mark keepers actually use is the eye-stripe — northern frontosa carry a thin, linear stripe down the face, whereas gibberosa wears a thick, triangular "Zorro mask" over much of the eye. In the seven-banded Kigoma form specifically, that ocular bar broadens toward its base and forms a triangle running from the eye to the mouth and the preopercle, a detail keepers use alongside the extra body bar to confirm the form. Selectively bred ornamentals such as the reddish "copperband" also circulate.
Hobbyists track wild frontosa by collection point — the exact stretch of shore a line was netted on — rather than by species, because the barring and the depth of blue shift from place to place and keepers avoid crossing localities. The genuinely Cyphotilapia frontosa northern points fall into two informal groups: a widespread six-banded "Burundi" type (Bulu Point, Karilani Island, Kavala, and the northern Congo shore sold as "Zaire North") and the seven-banded form restricted to Kigoma and neighbouring Bangwe on the eastern shore. The intensely blue southern lines traded as Blue Zaire (Kapampa, Moba, Kitumba, Tembwe), Mpimbwe, Samazi/Bismarck, Ikola, Kipili and the Zambian Isanga and Kasanga are, despite the "frontosa" on the bag, Cyphotilapia gibberosa — that species' account carries the full southern collection-point map.
Habitat
Cyphotilapia frontosa is endemic to Lake Tanganyika and restricted to its northern part, living allopatrically with its southern sister species Cyphotilapia gibberosa — the two ranges do not overlap. Seriously Fish bounds it roughly between Cape Tembwe on the western coast and Bulu Point on the eastern, taking in shoreline in the Democratic Republic of the Congo, Burundi and Tanzania.
It favours rocky, sediment-influenced slopes that fall away into deep water, with boulders broken by patches of open sand. How deep it actually lives is one of the more interesting arguments in the literature. Pierre Brichard reported vast schools — up to about a thousand fish — at 98–131 ft; Seriously Fish calls it most common between about 15 and 230 ft; yet gill-net data in the IUCN assessment put it most abundant much deeper, at 197–394 ft, with body size tending to rise with depth. The reading that squares these is a fish centred in deep water that moves shallower to feed, the biggest animals holding deepest. Its native water is hard and alkaline — pH around 8 or above — and warm but remarkably stable, the sheer volume of the lake buffering it against the weather.
One notable co-occurring species is the scale-eating cichlid Plecodus straeleni, which in Burundi and Congolese waters wears a coloration close enough to frontosa's own that it has been proposed as a case of mimicry, letting the smaller predator mix unnoticed with frontosa schools. The theory is not fully settled: field observers have reported watching P. straeleni swim among frontosa shoals without ever seeing it actually feed on them, while documenting it attacking other cichlids such as Cyathopharynx furcifer and Neolamprologus brichardi instead — so whatever the coloration is for, it may not be a straightforward case of "disguise to steal scales from this particular host."
Feeding
Frontosa is better described as a generalized predator than a specialist. FishBase and the IUCN both have it feeding on fish and macro-invertebrates, with mollusc shell fragments in the gut and some algae and detritus taken in — probably by accident as the fish mouths sand and rock. Its trophic level is modest for a predator (FishBase estimates about 2.9), which fits a mixed diet rather than pure piscivory.
The part keepers love to repeat is its hunting strategy: at dawn frontosa move up the slope after the open-water "sardine" cichlids of the genus Cyprichromis and the lake's clupeid sardines (Limnothrissa miodon, Stolothrissa tanganicae) and other small fishes, taking prey that are resting or poorly oriented in the dim light. A large mouth and an unhurried, deliberate approach suit an ambush hunter that overtakes slow or sleeping fish rather than chasing them down. Between such feeds it works the substrate for invertebrates. As a big, comparatively long-lived predator that gathers in large aggregations, it sits near the top of the rocky-zone community.
Mating
Frontosa is gregarious rather than strongly territorial in the way many rock cichlids are. Instead of each male holding a private patch, the fish form colonies organized around a dominance hierarchy: one or more alpha males claim the best shelters, and subordinate males fill in alongside a parallel female hierarchy. Aggression is genuine but diffuse — chasing flares around the dominant fish and sharpens at breeding — which is why the dominant male wants the largest caves while holding females need smaller refuges of their own. Both the literature and experienced keepers land on the same conclusion: group size and space, not pairing off, are what hold the colony's peace.
Breeding
Reproduction is maternal mouthbrooding, and it runs unusually long. Females lay small clutches of large eggs — commonly cited as 10–50, depending on the female's size — take them into the mouth right after fertilization, and carry the developing young for an exceptional 35–50 days or more. The striking part is documented science rather than folklore: Yanagisawa and Ochi (1991), working in the lake, found that brooding females go on feeding while they hold, running at about a quarter of the gut fullness of non-brooding adults, and that the larvae feed on particles inside the mother's mouth — "buccal feeding" — while still carrying large yolk reserves. Kuwamura (1986) likewise recorded females brooding young up to about 0.5 in, and an earlier comparative study by Balon put the incubation period at more than 50 days, with fry released at around 1 in — broadly consistent with the 35–50-plus days reported by Yanagisawa and Ochi. That long spell of in-mouth provisioning reads as an adaptation to a habitat with few safe crevices: the fry are not let go until they can fend for themselves.
In the aquarium
Rewarding but emphatically not a beginner's fish, and not one for a small tank — honest sources all agree. Live specimens were first exported from Burundi waters in the late 1950s, and the species has been a mainstay of the Tanganyikan-cichlid hobby ever since.
Tank: adults are big and the colony is the whole point, so footprint matters more than gallons. A 6 ft (71 in) tank is the widely cited practical minimum for a group; serious keepers run 7–8 ft (82.5–94.5 in), and Seriously Fish suggests dimensions around 240 x 60 x 23.5 in. Stock one or two males with five or six females, give the dominant male the largest caves and holding females their own smaller refuges, and the extra females plus hiding places keep bullying down.
Water: mirror the lake — hard, alkaline (pH about 8.0–9.0), warm and stable. Frontosa are notably intolerant of declining water quality, so they should never go into an immature tank, and weekly water changes are standard. As deep-water fish they like subdued lighting and dark corners and can spook under sudden brightness; ramp LEDs gradually and leave shaded refuges.
Feeding: high-quality pellets, mysis and frozen foods with a little vegetable content. Two cautions recur among careful keepers — avoid mammalian or avian meats like beef heart, whose fats the fish handle poorly, and skip feeder fish, which add parasite risk for little return.
Tankmates: frontosa ignore anything too big to swallow but will eat small fish, so pair them with robust Tanganyikans — Altolamprologus, larger Synodontis catfish, open-water cichlids — rather than tiny shell-dwellers. Two last realities from the trade: wild fish collected from depth must be decompressed slowly over days to avoid barotrauma, and the species is slow to mature, so a colony can take years to settle and spawn.
Conservation
The IUCN Red List reassessed Cyphotilapia frontosa as Near Threatened in 2025 under criterion A2d, a step down from its 2006 listing of Least Concern. It is still common and widespread within its deep-water range, but the assessment suspects a decline of roughly 20% over the last three generations — about 15 years — driven mainly by over-exploitation. The fish is wanted on two fronts: as a food fish caught locally for its taste and as a high-value ornamental export, with the Congolese colour morphs in the highest demand.
The threats reach past fishing. Sedimentation and erosion from shoreline development — road construction along the Burundian coast is singled out — degrade the rocky habitat the species depends on, and lake fishers already report localized declines near heavily fished areas. No targeted conservation program exists, though the species occurs in at least one protected area (Nsumbu National Park in Zambia) and would gain from the seasonal fishing closures the Lake Tanganyika Authority has trialed. For a slow-maturing, deep-living endemic with naturally low reproductive output, steady collection and habitat loss can quietly outpace recovery.