Heiko Bleher begins a two-part review of this fish of legend by examining the Serrasalmus genus.
It’s a complete myth that piranhas are man-eaters. I’ve made more than than 400 expeditions to South America and no-one has told me of any attacks on humans there.
Piranhas are found naturally east of the Andes, from the northern drainages to northern Argentina and occur in almost every type and body of water there. They belong to the Serrasalmidae family, which includes Pacus and Silver dollars.
This family is divided in three subfamilies, according to the late Jacques Géry: Myleinae (Pacus and others, Catoprioninae (Wimpel piranhas) and Serrasalminae (the carnivorous piranhas).
All have a compressed body, usually deep and almost disc like. Spines, called serrae, can vary from six to 70, and serrae gives the origin of the scientific name of family, subfamily and the generic name Serrasalmus — to which Géry had placed all species.
He divided the Serrasalmus genus into five subgenera:
There’s Pygopristis with a single species identified by symmetrical teeth, usually with five cusps; base of the anal fin covered with no more than two scale rows and no teeth on the palate.
Pristobrycon has at least four species identified by only three or four teeth on an obtuse palate, eventually lacking any.
Serrasalmus has eight to ten on a well-formed palate suborbital covering at least three-quarters of the cheek.
According to Géry, only Pygocentrus and Taddyell are true piranhas. Both have broad, heavy heads and jaws, flat snout and prominent mandible. The only difference is the presence of rudimentary rays in the Pygocentrus adipose fin.
Piranhas will only attack in self-defence and are the cleaners of South American rivers, usually eating dead, dying and weak fish. They eliminate natural meat waste.
A total of 44 species should be considered valid, but I have collected and recorded many more different forms.
In this first part of the guide I'll highlight the Serrasalmus genus.
Linnaeus described the first piranha in 1766 as Salmo rhombeus from the type locality of Brokopondo, Suriname river, Suriname and many pictures since of different piranha are still described as S. rhombeus.
The Catalogue of Fishes and Fishbase.org list its distribution as the Amazon and Orinoco river basins, northern and eastern Guiana Shield, and north-eastern Brazilian coastal rivers: Argentina, Bolivia, Brazil, Colombia, Ecuador, French Guiana, Guyana, Peru, Suriname and Venezuela.
Old photographs reveal S. rhombeus juveniles as having large black spots, a black band on the caudal base and large black caudal seam at the caudal end and anal fin. The species is almost pointed at this stage, and as an adult grows to about 35cm/14” TL.
I’ve spent much time researching piranha with Géry, collected throughout their range and I‘ve separated these good species:
Serrasalmus niger (Jardine in Schomburgk 1841) must be valid but placed into Pygocentrus. It’s greyish and bulky with almost black fins and totally different from type rhombeus from Surinam. The typical white black-banded tail is absent, the snout not pointed and its average length is 18cm/7” SL.
Serrasalmus immacolatus (Cope 1878) — pictured above — has a type locality in the Peruvian Amazon. Cope described a fish without pattern and called it "immaculate". His description and drawing matches piranha I found in the Peruvian Amazon, as well as in the Rio Japurá, Rio Juruá, Río Calderón and it seems restricted to the upper Amazon basin. This is a high bodied, black piranha. that’s also one of the largest, at about 40cm/16” SL.
Serrasalmus rhombeus ssp., as mentioned by Géry from the Araguaia, Tocantins and Xingu basin is similar to the type, but different in size, colour pattern and possibly a species by itself from south of the lower Amazon. I classified the latter for the time being under Serrasalmus sp. 9. These are S. cf. rhombeus and S. aff rhombeus (pictured above) in my guide.
My Serrasalmus sp. 4, sp. 5, sp. 6, sp. 7, and sp. 8, are all from different collecting places, but grouped by some under S. rhombeus. However, these are all different in morphology and possibly still nameless.
One of the most beautiful is S. spilopleura (above). Those I collected at the Guaporé river are bright yellow from the lower mouth to anal fin. The body is a brilliant blue, tail end and the dorsal and seams jet black with yellow stripes. However, all bar its brilliant scales are lost in elder specimens.
Juvenile piranhas of different species may have a similar, often spotted colour pattern, which can confuse. The best way a layman can identify a species is when the fish is semi-adult or a year old.
S. auriventris and S. nigricauda of the Serrasalmus species, described by Burmeister in 1861, are doubtful, having never been seen alive and descriptions are poor. There’s also confusion between S. cariba, collected and described by Humboldt, and Pygocentrus notatus which is a small shoaler in Venezuela. S. cariba has no humeral spot, while P. notatus, placed in synonomy to S. cariba, always has a large black humeral spot and is much smaller.
Serrasalmus elongatus is also confused and several species are under this name. All are more or less elongated, but the one Kner described in 1860 from the Guaporé river, is the least so and often mis-identified.
The Venezuelan species named S. elongatus, is the described Serrasalmus pingke (Fernández-Yépez 1951), from the Río Apure. I found a very similar one in the Rio Guaporé and males have red bellies (pictured above).
Serrasalmus humeralis (pictured at the top of the page) is the largest piranha I’ve collected, with specimens more than 45cm/18” TL taken from the Xingu.
Serrasalmus iridopsis (Cope 1872), from the type locality Río Ambyiacu, upper Amazon was recognised by Géry as a good species, but recent workers placed it as Serrasalmus humeralis (Valenciennes 1850). Géry, however, said it is very distinct from the latter.
Then there’s Serrasalmus maculatus from the Guaporé river. I collected two variants in one spot, but in 2010 the upper specimen was separated, according to molecular evidence, and newly named S. odyssei (pictured above).
S. manueli is another problem. That described from Venezuela is red around the breast up to semi-adult, while the one I collected in the upper Aracá river (Rio Negro basin) is always black. Both have a large vertical humeral spot similar to the S. humeralis from the Xingu. The specimens I collected in the Rio Negro again are different, but S. manueli belongs to large piranhas species, measuring 45cm/18”
The species described in 1993 as Serrasalmus neveriensis from the Río Neveri basin, in a coastal area of Venezuela, is very similar to S. medinai. I collected what must be the same species in the Rio Negro basin and a similar form in the middle Purus basin as well — my Serrasalmus aff. neveriensis.
Another species, described in 1969 as S. nalseni, comes from the Orinoco basin and also has spots, but only it seems until semi-adult. This has been often confused with similar species.
How to keep Serrasalmus
Your tank décor should consist of white or beige sand, soft substrate with leaves, driftwood and/or rocks — creating places where piranha can hide or escape to.
It’s not easy to keep them with plants, but include some floating ones to create shade. Any piranha aquarium must be at least 80cm/32” long, even for the smallest species. For larger species, 100cm/40” is the minimum length.
Always keep shoaling piranhas in groups, like young S. gouldingi and S. manueli. Species like S. hollandi, S. maculatus, S. serrulatus or S. spilopleura should be kept in at least threes. Keep S. elongatus, S. pingke, S. nigricans (pictured above) and S. niger as individuals. These live singly in nature, except when mating.
Every piranha aquarium should have excellent filtration and periodic partial water changes — especially if food is left over. I suggest 20-30% weekly.
Water parameters depend on the species, but most are used to nearly neutral water conditions, from pH 6 to 7.2, even 7.5; conductivity 50 to 300 µS/cm. Temperature should always be above 24°C/75°F — the best average being 27-28°C/81-82°F. Some piranha species may jump too, so include a tank cover.
A biotope correct aquarium should contain only the South American fish your piranha already knows, and for only as long as you can guarantee that the piranhas are fed daily.
Although piranhas cannot cannot devour armoured catfishes, they will still try — in the picture above they have eaten the top fin of this Hypostomus.
Serrasalmus live with freshwater stingrays, with groups of Myleus and Metynnis, some large Hypostomus and Pseudacanthicus, and also with Heros, Astronotus, Hypselecara, large Mesonauta, Cichla and large Crenicichla, adult Boulengerella and Chalceus, as well as giant Leporinus.
A nature-correct choice of tank mate should ensure harmony for everyone. However, an ideal length of aquarium for such a community would be 250cm/ 8’ — so bear these requirements in mind if considering taking such a route.
Meet the species
There are 27 recognised Serrasalmus, with four more awaiting recognition.
The species numbered 4 to 9 also need to be named, in addition to those still grouped under S. rhombeus (with cf. and aff. as mentioned).
Merckx, Jégu and Santos 2000. Type locality: Pitinga river, Uatumã basin, Amazonas, Brazil. Distribution: Amazon river basin, Uatumã river, Brazil.
Ramírez 1965. Type locality: El Polvero, Río San José, Guárico, Venezuela. Distribution: Orinoco river basin.
(Burmeister 1861). Type locality: Río Paraná, near the Quinta Argentina. This is only known from Argentina.
Lütken 1875. Type locality: Lagoa Santa, Minas Gerais state, Brazil. Distribution: This species is only known from the São Francisco river basin, Brazil.
(Humboldt 1821). Type locality: Apuré and Orinoco rivers, South America. No types known. Also a ‘Red belly piranha’ without humeral black spot and should be placed into Pygocentrus.
Jégu, Leão and Santos 1991. Type locality: Laguna Mocovi, Río Mocovi, Béni province, Bolivia. Distribution: Middle Amazon river basin: Bolivia, Brazil and Peru.
Norman 1929. Type locality: Rockstone, Guyana. Distribution: Amazon river basin and northern and eastern Guiana Shield rivers: Brazil, French Guiana, Guyana, Suriname and Venezuela.
Kner 1858. Type locality: Rio Guaporé, Mato Grosso, Brazil. No types known. Distribution thought to extend throughout the Amazon and Orinoco river basins. Picture shows drawing from 1860.
Jégu and Santos 1988. Type locality: Rio Araguaia, Araguacema, Brazil. Distribution: Tocantins basin.
Castelnau 1855. Type locality: Rio Araguay, Brazil. Distribution: Tocantins basin, Brazil.
Fink and Machado-Allison 1992. Type locality: Anavilhanas, Río Negro, Brazil. Distribution: Amazon and Orinoco river basins.
Fink and Machado-Allison 2001. Type locality: Rio Branco, Marara beach, Brazil. Distribution: Negro river basin, Brazil.
Eigenmann 1915. Type locality: Rio Guaporé at Maciél, Brazil. Distribution: Madeira river basin.
Valenciennes in Cuvier and Valenciennes 1850. Cited type locality: Araguay river. Distribution: Restricted to Xingu basin.
Cope 1878. Type locality: Peruvian Amazon. in places in synonymy with S. rhombeus. This high-bodied black is restricted to the upper Amazon basin.
Cope 1872. Type locality: Río Ambyiacu, Peru. Some place this in synonomy with S. humeralis, but this is different to the large S. humeralis from the lower Amazon. Distribution: Only known from the upper Amazon basin.
Peters 1877. Type locality: San Fernando Apure, Venezuela. Distribution: Orinoco river basin, Venezuela. This is Also found in El Polvero, Río San José, Estado Guárico, Venezuela.
Kner 1858. Type locality: Rio Guaporé, Mato Grosso, Brazil. Distribution: Amazon and Paraguay-Paraná river basin: Argentina, Bolivia, Brazil, Colombia, Paraguay, Peru and Uruguay.
(Fernández-Yépez and Ramírez 1967). Type locality: Río Paraguaza, Middle Orinoco river, Venezuela. Distribution: Amazon and Orinoco river basins: Brazil and Venezuela.
Valenciennes1837. Type locality: Rio Paraná, Brazil. Distribution: Throughout the Paraguay-Paraná river basin: Argentina, Brazil, Paraguay and Uruguay.
Ramírez 1965. Type locality: El Polvero, Río San José, Guárico, Venezuela. Distribution: Orinoco river basin, Venezuela.
Fernández-Yépez 1969. Type locality: Río Uracoa, Estado Monagas, Venezuela. Distribution: Orinoco river basin, Venezuela.
Machado-Allison, Fink, López Rojas and Rodenas 1993. Type locality: Río Querecual, tributary of the Río Neveri, Querechual, Estado Anzoálegui, Venezuela. Distribution is in the coastal rivers of Venezuela.
Jardine in Schomburgk 1841. I believe this is a valid species. Please refer to Pygocentrus in part two of this guide (coming soon).
Spix and Agassiz 1829. Type locality: Equatorial rivers of Brazil. No types known. Distribution: Amazon river basin, Brazil.
Burmeister 1861. Type locality: Río Paraná, near Rosario, Argentina. No types are known. Distribution listed as Argentina.
Hubert and Renno 2010. Type locality: Rio Itenez, Bella vista, Madeira, Bolivia. Distribution: Rio San Martin and Rio Paragá, Bolivia.
Fernández-Yépez 1951. Type locality: Río Apure, La Defensa, suroeste de San Fernando de Apure, Estado Apure, Venezuela.
(Linnaeus 1766). Type locality: Brokopondo, Suriname river, Suriname. The ‘real’ S. rhombeus is restricted to the northern part of South America, from the Orinoco river basin east to Suriname.
Géry1964. Type locality: Caño Yarina, Río Pacaya, tributary of Puinahua canal, branch of lower Río Ucayali, Peru. Distribution: Ucayali river basin, Peru.
(Valenciennes in Cuvier and Valenciennes 1850). Type locality: Amazonas, Brazil. Distribution: Amazon river basin, Peru, Argentina.
Kner 1858. Rio Guaporé, Bobota, Mato Grosso, Brazil. Distribution: Guaporé river basin, Paraná river basin, Argentina and Brazil.
Serrasalmus sp. 4
Type locality: Upper Rio Guaporé near Vila Bela, Mato Grosso. Claimed variant of S. rhombeus, but morphology is different.
Serrasalmus sp. 5
Type locality: Rio Jarí merging into Lago Jari, Purus basin, Amazonas region of Brazil.
Serrasalmus sp. 6
Type locality: Rio Iriri, land locked, above a waterfall, Xingu basin, Brazil. This is a schooling piranha, unlike S. rhombeus or other single large piranha.
Serrasalmus sp. 7
Type locality: Caño Mirim, affluent of Rio Yavari in Peru.
Serrasalmus sp. 8
Tye locality: West of Abuná in the Rio Abuná, Acre, Brazil.
Serrasalmus sp. 9
Type locality: Rio Araguaia north of Aruaná near Ilha do Bananal, Tocantins, Brazil.
Why not take out a subscription to Practical Fishkeeping magazine? See our latest subscription offer.