Definitive guide to elephantnoses


Heiko Bleher looks at the Mormyridae family of fish that locates and communicates by using its own electricity.

There are more than 190 valid species of elephantnoses in the Mormyridae family and they belong to the Osteoglossiformes order.

Their distribution is confined to the African continent, but I've found from the lower Nile to South Africa in almost every type of water.

They are mostly night active and best known for the species with elongate body forms or tube snouts — and for the organ they use to generate electric fields to help them navigate, detect prey and communicate in a process known as electro-location.

The South American Gymnotiformes, including the Black ghost knifefish, also have electric organs that operate in a very similar way. Some knifefishes even have a similar head form, arealso night active and live in very similar habitats. However, they are not thought to have shared any recent ancestry.

The elephantnose’s specialised organ is in the caudal peduncle which discharges the pulses. Each one builds up an electrical field around the fish and is sensed by cutaneous (skin) electro-receptor organs distributed over most of the body surface.

Nearby objects distort the field and cause local alteration in current flow in the electro-receptors closest to the object. By constantly monitoring responses of its organs, a fish can detect, localise, identify objects and communicate with others up to about 3m/10’ away.

When the fish is resting discharge frequency is low, but this rapidly increases to between 80 and 100 pulses per minute when disturbed. However, the strength of electricity created, about one volt, is not enough to seriously shock a human.

Mormyrus rume is an exception, giving off 25 volts. Its discharge can be felt by touching the caudal peduncle of the fish when taken out of the water.

Covering the genera

There’s a total of 21 genera in the Mormyidae family and here I’ll cover the genera and species found in the hobby, imported or occasionally bred.

Mainly because of their often cryptic habitat and because they are night active, there are still many undescribed and unknown species.

The known genera are: Brienomyrus, Brevimyrus, Boulengeromyrus, Campylomormyrus, Cyphomyrus, Genyomyrus , Gnathonemus, Heteromormyrus, Hippopotamyrus, Hyperopisus, Isichthys, Ivindomyrus, Marcusenius, Mormyrops, Mormyrus, Myomyrus, Oxymormyrus, Paramormyrops, Petrocephalus, Pollimyrus and Stomatorhinus.

Did you know?

Elephantnoses even amazed the ancient Egyptians! We know from records from 2450-2350 BC that many carvings were made of them and still these survive, as in the tomb of Idout, tomb of Ti and tomb of Mereruka, and also in some beautiful paintings.

To that dynasty of Egyptians these fishes were sacred, a symbol of fertility and one of the forms of the goddess Isis is depicted with one in her hand.

First records

The first scientific record of an elephantnose comes from Forskål in 1775 with Mormyrus kannume, followed by Linnaeus (1758) with three more species (cyprinoides, anguilloides and cashive – pictured above), also from the Nile. The fifth Nile species, niloticus, was published in 1801 by Bloch and Schneider.

These were mostly those already known from ancient paintings and carvings. It took almost another 50 years before new elephantnoses were described and only Peters, in 1852, mentions the first from Mozambique and Castelnau, in 1861, another from the Zambezi river.

Others followed, but Boulenger was the specialist, describing between 1890 and 1920 almost 60 new species from all over Africa and which are still valid today.

Last year five new species of Pollimyrus were described from a very small area in the Odzala National Park of the Democratic Republic of the Congo (DRC) in a recent publication by Lavoué, Sullivan and Arnegard. This shows how many species are still undiscovered in Africa.

Different snouts

These fishes live in amazing habitats and have evolved extraordinary adaptations among individuals found in a completely different location of the same river.

Each snout is specialised to the nature of the river floor in which it feeds. Long and thin ones allow fish to probe for food in deep and small-grained gravel or sand, while short and fat ones enable them to feed on algae-caked bedrock.

The species Gnathonemus petersii from the lower Niger basin, was one of the first to arrive in the hobby and it remains the most popular. It’s also the reason that the common name of elephantnose evolved.  

Its chin is prolonged as a fleshy, sensitive finger to search and find food in the soft ground and between small stones, while the mouth remains in the normal position.

However, species of the genus Campylomormyrus, for example, have the snout extended into a long downwards-curving tube. In this case the small mouth is at the end of the tube.

Not all elephantnoses have elongated snouts. Petrocephalus species (above) have an almost bulldog-like appearance.

Although they have poor eyesight, they have extremely large brains, between one-fifth and one-eighth of their body weight. This is more than 25 times the relative weight of a pike’s brain and approaches the human ratio of about one twenty-fifth to one-fortieth. Their cerebellum is greatly enlarged and this is related to the perception of the fishes’ electric field.

It’s uncertain when the first elephantnoses arrived in the hobby but we identify Gnathonemus petersii (above) with the family and it is widely found and exported from Nigeria. The other Gnathonemus species have hardly been exported. Gnathonemus can grow to maximum of 25cm/10” TL.

The species of the genus Brienomyrus are rarely more the 12-15cm/ 4.8-6". B. longianalis is exported from Lagos, Nigeria, as are three unidentified species: B. sp. 1, B. sp. 2 and the undescribed B. sp. 3 from the lower Congo. This is in addition to B. sphekodes from the Doumé region in the Congo and B. tavernei from the Tumbwe river system in the DRC  which are rarely exported from Kinshasa.

The two species, or variants, of the genus Brevimyrus are infrequent in the hobby, but the Niger species pictured above, originally from Gambia, has also been exported from West Africa together with Brevimyrus sp. 1, from the lower Niger river delta, in Nigeria. They remain around 10-12cm/4-4.8” TL.

The monotypic Boulengeromyrus knoepffleri, with its amazing snout, is from Gabon, recorded at a maximum 30cm/12” and a hobby rarity.

On the other hand, the species of Campylomormyrus is probably the most colourful elephantnose, at least in the juvenile stage. Some grow to 60cm/24” and are often colourless at that size. They are confined to the Congo basin, are popular in the hobby, but infrequently exported.

The species C. alces (above) C. cassaicus and C. christyi, are spasmodically exported from Kinshasa, as is C. elephas and C. cassaicus. However, the very long-snouted C. curvirostris, C. numenius and C. tshokwe rarely arrive in the hobby.

To my knowledge the species of the genus Cyphomyrus was never imported and the monotypic Genyomyrus donnyi (pictured above) rarely so. The latter is easy identified by its unique lower snout.

I have collected them several times, but brought back only a few specimens from the upper Congo and it was difficult to reach the upper Lufira waterfall where they’re found. They are solitary fishes and the largest recorded was 45cm/18”.

The monotypic genus Heteromormyrus from Angola has never been in the hobby.

Members of the genus Hippopotamyrus are rare, but H. aelsbroecki and H. weeksi from the Congo basin, are seen more than H. discorhynchus from the rivers in the southern Lake Malawi region in Mozambique, they grow hardly longer than 12-15 cm/4.8-6”.

There are also monotypic species from the Nile and the mostly desert regions of Mali, Burkina Faso and central Niger River regions, where several subspecies have been recognised.

Hyperopisus bebe, which in most places is a food fish, is rarely in the hobby. It grows to 50cm/20” and has hardly any colour.

However, the small monotypic Isichthys henryi is beautiful and a very different elephantnose species and excellent for smaller aquaria. It is occasionally found — unlike, to my knowledge, the two described species of the genus Ivindomyrus from Gabon and DRC.

The Marcusenius genus (pic shows M. monteiri) was named after ichthyologist Johann Marcusen who published extensively on elephantnoses. Its species have a short and blunt snout, their mouth is terminal and a fleshy prominence extends ahead of the lower jaw. They are found in most parts of Africa and 35 species are recognised valid.

Most common in the hobby is M. cf. macrolepidotus. Growing to 25cm/10” it has a nice marble colour and looks like the described species from Mozambique.

Also shipped out regularly are M. cyprinoides (above) from the lower Niger: M. senegalensis, of which I found two races in the Niger, as well as M. sp. 1, (still unidentified).

The most beautiful is M. moorii (above) which I found in southern Cameroon but is rarely exported. Also desirable is M. livingstonii from Lake Malawi, but it’s not found in the hobby.

Mormyrops have the largest fishes in the family, with specimens collected at 1.5m/5.2’. My largest was near Limala, on the Congo in 1986, and that was M. curtus at 130cm/4.3’. Mormyrops are all elongate to anguilliform and have a terminal mouth. They have strong, truncate or notched teeth and can be very aggressive. Hobbyists should avoid M. curtus and M. anguillioides (pictured above).

My only suggestions for aquarium life would be the small and peaceful M. parvus and similar looking M. sirenoides.

Species occasionally found from the Congo are M. lineolatus which I found in the upper river and M. engystoma from the lower basin. They remain between 15-20cm/6-8” and need hideaways in the aquarium. If kept in twos they could fight.

The Mormyrus genus with 22 species is easily recognised by very long dorsal fins and very short anal fins. Most have a tubular snout and largest specimens can grow to 45cm/18”. Few reach the hobby, mostly M. rume from Nigeria and M. caballus from the Democratic Republic of Congo (DRG). M. caschive and M. ovis from above Lufira waterfall.

M. macrophthalmus, is a strange looking fish, hardly exported, and even less so is M. hasselquisti from the middle Niger. None have shown to be aggressive, but keep three or more together.

The two long-snouted Oxymormyrus species (picture shows O. boulengeri) are small and unique, hardly more than 20cm/8”, but hardly found in the hobby. They are also difficult to collect from inaccessible crevices.

Almost all the seven known Paramormyrops species are restricted to lower Guinea, except for P. kingsleyae, and rarely found in the hobby. Most are less than 10cm/4” and have a non-tubular snout, often bluntly rounded or tapering to a gentle point. The mouth is small and terminal.

Some of the 28 species of Petrocephalus can be found in the hobby, but only P. ansorgii and P. pallimaculatus from the Niger river are shipped from Lagos. The Congo species, P. simus, with silver body and black spot below the dorsal fin and P. cf. microphthalmus (similar to P. ansorgii), are rarely exported.

Petrocephalus are also small, rarely over 10cm/4” and peaceful.

There’s 17 small species from Pollimyrus and almost all will suit a biotope correct aquarium.

There’s only four or five in the hobby, but P. isidori from the Niger, has been bred many times. P. adspersus and P. isidori are frequently exported from Lagos.

P. lhusyii is beautiful, hardly 8cm/3/1”, so ideal for smaller aquaria.

Stomatorhinus has never been in the hobby. these are 4-10cm/1.6-4” long.

The relatives…

Also in the Osteoglossiformes order with elephantnoses are the Arowana (Osteoglossidae), Pirarucú (Aprapaimidae), butterflyfishes (Pantodontidae), mooneyes (Hiodontidae), knifefishes (Notopteridae) and Aba aba (Gymnarchidae).

How do I keep them?

Don’t place them in a small aquaria. Each tank should be at least 100cm/40” long, and the bigger the better.

As for décor this depends on the species. Gnathonemus petersii, the one most frequently seen in the hobby, needs fine sand in at least some part of the aquarium. This species must be kept in groups of at least five, as they live only in groups —which is not the case with most of the other elephantnose species.

The aquarium also needs some gravel and, if you wish, some larger stones and well-watered driftwood and leaves. There are rarely plants in the natural elephantnose habitat, but go for Crinum natans, Bolbitis heudelotii and Anubias species if wanting greenery.

The larger species should have larger tanks and be kept in pairs or groups of three or more. No elephantnose fish should be kept alone.

Aquariums for any of the Marcusenius species should be at least 100cm/40” long and for Mormyrus at least 150cm/60”  — likewise the other larger species.

What food should I feed them?

The mouths of most elephantnoses are small with a few weak teeth, so food should be offered accordingly. In nature most feed on aquatic insects, their larvae and snails. The very large ones also take fishes.

In aquaria offer Rotifera (rotifers), Cladocera (water fleas) and Copepoda (copepods) with a variety of foods. I have often seen elephantnoses become very skinny in importers, shops and private aquaria and die quickly because of not enough food or the correct nutrition.

They need mosquito larvae, bloodworms, white worms, black worms, Daphnia, Tubifex and Cyclops, if not live then at least frozen. If you can’t get hold of these, or are unable to offer them every day, don’t keep elephantnoses!

Can I spawn them?

Spawning is not as difficult as it seems. Frank Kirschbaum proved it with four species from four genera by simulating the high water season, by raising the water level and reducing conductivity to imitate rain.

He also proved that mormyrids kept under such conditions for two to four weeks will attain sexual maturity and begin to spawn.

Regular spawning at intervals of several days follows, with length of interval varying according to the species. It’s five to eight days for Pollimyrus isidori, 20 to 30 for Mormyrus rume and between 30 and 50 days for Campylomormyrus cassaicus (pictured above)

These mormyrids are typical fractional spawners and their period of reproductive behaviour continues, sometimes for many months, until low water conditions (a rise in conductivity) trigger a cessation of activity.

After three to four weeks the sexual organs become inactive and the fishes await the next high water to trigger further ripening of their gonads — which was the case in nature when I saw them spawning during the rise of water. One example is Pollimyrus isidori, which can become sexually mature at 6cm/2.4”.

To simulate the rainy season get an aquarium of at least 120 x 50 x 60cm/48 x 20 x 24”, as the male will already occupy a territory of 50-100cm/20-40” across. Conductivity should go from 2,400 µS/cm slowly down to 200, water level from 25/10” to 45/18” or 50cm/20” high and pH from about 7, dropping to below 5.

The male will build a nest of plant material in his occupied territory, so make some available. When spawning the ripe female invades his territory and deposits her eggs on the substrate near the nest. The male gathers them — and 80 and 200 may be laid in a single night — puts them in the nest where he guards and tends them.

The embryos hatch on the fourth day and the larvae start to take Artemia from about the thirteenth day. They remain under guard near the nest for some time.

During courtship, acoustic signals created by the male are used for communication between sexes.

It was similar with Mormyrus rume (pictured above) which were sexually mature at about 15cm/6”,

Camphylomormyrus cassaicus were best, with a ratio of two males to three females in an aquarium at least 200cm/80” long and were also sexually mature at 15cm/6”.

Hippopotamyrus pictus also bred this way.


For most Congo basin species you could add Distichodus sexfasciatus, D. noboli, Phenacogrammus interruptus, P. caudalis, P. caudomaculatus Ctenopoma acutirostre, Synodontis contractus, S. nigriventris, S. flavitaeniatus, Pantodon buchholzi and possibly Hemichromis stellifer and Xenomystus nigri.

For Nigeria species try Pantodon buchholzi, Brycinus longipinnis, B. chaperi, Arnoldichthys spilopterus Ladigesia roloffi (at least five), Microctenopoma ansorgii and Gnathonemus petersii (also five).

Aquarium operates by fish power

The features for electro-location, as used by elephantnoses, are so ingeniously constructed that researchers sought to mimic their capabilities and, by the 1980s, experts at Aquarium Nancy in France had all their computers run by the electric discharge of Gnathonemus petersii — the elephantnoses I had supplied to them.

The big clock at the entrance of the aquarium has been powered for 20 years solely by the discharge of a large Gymnarchus niloticus.

Mystery of the flashing water

I once had an incredible night experience with elephantnoses in the Lake Mai Ndombe region of the Democratic Republic of Congo. While out collecting in the flooded rainforest in some 50 cm/20” of clear water, my flashlight batteries ran out.

It was pitch dark and I looked in the water where I was standing and suddenly it started to blink around me. Near my legs was a group of five or six elephantnoses swimming over fallen leaves and tree branches.

Why could I see them? Their eyes were flashing on and off in bright (infra-red?) colours and, while watching this for some 15-20 minutes, I began to realise that they were using some form of communication, like Morse signals.

At least one pair was mating at the same time, twisting around each other. It was an amazing sight — a spectacle I’ll never forget. 

Check out Heiko's other definitive guides in this series:

Discus part one

Discus part two


South American freshwater rays

Why not take out a subscription to Practical Fishkeeping magazine? See our latest subscription offer.