Contribution to the biology of the Pirarucu, Arapaima Gigas – (Cuvier) in captivity, (Actinopterygii, Osteoglossidae), Fortaleza, Ceara, Brazil, 1959. 15 p. plates. 23 cm. (Brazil. Servico de Piscicultura. „Ser. 1-C. Coletanea de Trabalhos Técnicos. ” English abstract. Publicacao no. 177)
„CONTRIBUTION TO THE BIOLOGY OF THE PIRARUCU.
ARAPAIMA GIGAS (CUVIER) IN CAPTIVITY
por Osmar Fontenele
— 399 —
I was encouraged to edit and divulge the various notes taken down during the several phases of the breeding of Pirarucu, Arapaima Gigas (Cuvier), at the Posto dé Piscicultura of Lima Campos, in the State of Gears, Brazil, by Dr. Carlos Esteváo de Oliveira, former Director Museu Goeldi, Belém, Pará (whose death occurred in 1946), to the memory of whom I dedicate this work, which I hope will further the knowledge of this valuable species of the Amazon River. Incidentally, it is well to quote the opinion of Eigenmann & Allen (1942):
„The most important Amazonian fish is one of the least known. Its life history, habits, food, migrations, are not only unknown, but misinformation about them prevails, and a fertile subject for future study awaits someone”.
I had planned, with the valuable help of Dr. Esteváo, some months prior to his death, to start a work, the most complete possible, on the Pirarucu, based on the facts he had gathered during manу years of close observation, at the Museu and in the region where this big fish has its natural habitat. Unfortunately, however, the death of Dr. Esteváo, whose vast knowledge of the Amazonian ichthyofauna would have been of great help, prevented me from widening the scope of this work, which is thus shortened to encompass only the biology of the pirarucu in captivity.
Among the various kinds of fish acclimated by the Serviçe de Piscicultura of the Departamento Nacional Obras Contra as Sécas, in dams scattered throughout Northeast Brazil, in its program of planting and replanting such water reservoirs, the pirarucu stands out by its big size, great precocity, easiness of capture and the possibilities it offers for a profitable industry.
Among this species are found the largest representatives of the freshwater ichthyofauna in the world.
The word pirarucu comes from the language of the Brazilian Indians and is formed by the Indian words: PIRA, meaning FISH, and URUCU, which means RED, thus the RED FISH, since it is the red coloration shown by the scales of certain regions of its huge body that makes itself immediately noticed.
The intensity of the scale coloration and number of such scales vary in accordance to the nearness of the spawning season. Observations on the habits of this species started at the Posto de Piscicultura de Lima Campos, from where, sometime later, thousands of fry were transplanted to six big Government dams in the States of Paraíba Rio Grande do Norte and Ceará.
— 400 —
It should be mentioned that the fish-ponds designed by Dr. Antonio Carlos Estevão de Oliveira were of the utmost importance in our work of gathering data for this paper.
The tanks are twelve, of gradual depths, ranging from 0.80 to 1.80 m with transversal and longitudinal sections, in trapezoid form, communicating one to the other by means of a 1,00 m² gallery with walls made of brick, the bottom being the natural earth. On the upper part the tank is 24 meters long and 6.30 m wide.
The reason for different depths and intercommunication was the lack of information on the suitable depth for breeding this species.
In the recesses at the top of each tank, a plant of Amazonian origin known as Canarana (Panicum sp.) was sown to shelter shrimps among its roots, since the pirarucu is very fond of this food in certain phases of its development.
Each tank receives water directly from a channel that flows alongside, by means of a gate that permits an easy control of the liquid. Cleaning is made by drainage at the last and deepest tank.
Observations started in 1942, on 19 pirarucus which came from the Amazon River, all born in 1939 and with the average lenght of 1.20 m. Those would be the future breeders of Posto de Piscicultura of Lima Campos.
One of the physiological particularities observed right at the start was the habit the fish has — when not disturbed — of often coming slowly up to the surface to make its suplementary breathing, in addition to its branchial breathing.
The fishermen take advantage of this habit to catch the pirarucu. Standing on the prow of their canoes, harpoon in hand, they await the fish coming to the surface of the water, snout foremost, open-mouthed, for the needed portion of air. As the fish goes down again, its dorsal region shows above the water surface. It is then that the fishermen expertly let go the harpoon to an easy target. At this phase of their development, the future breeders were fed dead and live fish. We noted particular fondness for a fish commonly known as cascudos (Loricariidae). The cascudos are caught through a strong suction made by the pirarucu, which causes a noisy movement in the water. All the water that goes inside when the fish swallows its catch is later expelled, forcing for this an exaggerated opening of gill covers.
In spite of its big size, the pirarucu has always seemed to us a harmless creature, as it is not equipped with spikes or sharp teeth. The maxillary of a 2.00 m long fish are provided with a row of few conical teeth, less than 2 mm long.
One curious anatomical characteristic of this fish is its bony tongue which has little mobility. This peculiarity gave the family its name — Osteoglossidae. Besides the aforementioned organ, there is the hard palate and two bony plates on the sides. This bony, jagged structure is strong enough to enable the fish to crush its prey before swallowing it. Together with the pharyngeal teeth, with which the fish la also provided, this jagged formation does the work of real teeth. The pharyngeal teeth, varying in number, have the form of jagged plates of little thickness and are located in different places of the pharynx.
— 401 —
After reaching over 1.70 m of total length, Pirarucus reared in captivity eat little and show large reserve of adipose substance.
One peculiarity that deserves special attention is the conformation and location of the reproductive organs of this species.
OVARY — Contrary to the near totality of other fish, the ovary of the female pirarucu is an odd organ, depressed, brittle, with elongated extremities, the outer end being sharper than the other, the part that adheres to the perytoneum being flat, with various contours, and the opposite side — the one that is turned downwards when the fish is in its normal position — free. The anatomical constitution of the ovary is leafy, like the leaves of a book. Between each two of these leaves — that have their edges soldered — are the ovules, tied to the stroma.
The ovary is located at the fore two thirds of the abdominal cavity, In the middle of the left side. Its extremity, at the tail, opens to the genital orifice, which has a diameter of about 6 mm, and is behind the anus, at a short distance from the latter. In a 1.90 meter long pirarucu, the ovary at oestrus stage, with 495 mm in total length, and 120 mm as maximum width, weighed 650 g (Protocol 205). The female sample, registered under no. 140, with 1.85 m in length, had a 521 mm long ovary, which weighed 450 g. Another 1.90 meter long female had an ovary that weighed 1.300 g (Protocol No. 275). This same protocol book registers, under n.° 280, a female which had a total length of 1.89 m, with the ovary weighing 1.200 g. Therefore, the weight of the ovary of female samples with an average of 2 m in length, ranged from 495 to 1.300 g.
The ovary coloration is given it by the ovules. In preparations, when fixed by formaldehyde at 5%, the ovary shows a light yellow coloring that is almost white.
However, when fresh at examination, the ovary at oestrus stage shows a variable coloring, the strongest hue being that of pretroleum green. The eggs show different dimensions, coloring, form and fixation, according to the stage of development and, consequentiy, the pirarucu is classed in the group of fish that have partial sexual maturation, what causes divided spawning.
The examination of an ovary, even at oestrus stage reveals the existence of light yellow and yellow ovules, the smallest being round, with up to 1 mm in diameter, strongly adhering to the stroma; medium-sized ovules light green in color, roundish, with diameters varying from more than 1 mm and less than 2 mm, adhering less to the stroma than the former; and ovules of dark green coloring (petroleum green), roundish wilth diameters varying from 2 mm and a little over 2.5 mm, which can easily be detached from the stroma.
The resistance shown by the ovules when one tries to separate them as well as their dimensions indicate that they have reached full maturity when they have taken the petroleum green coloration. We counted the ovules in four ovaries after they had
— 402 —
Efetuamos o cálculo do conteúdo de óvulos de quatro ovários, contando o número de óvulos existentes em 10 g, após terem sido os mesmos separados pelas dimensões atingidas. Foram ar, seguintes as médias dos resultados obtidos, em 10 g de ovário: a) óvulos maiores, maduros, de cór verde escura — 448; b), óvulos médios, de cór verde clara — 447; c) óvulos menores, de cór amarela, imaturos — 826.
Embora reconhecidamente pequeno o número de contagens efetuadas, podemos, todavia, com os cálculos realizados concluir que o número, em média, de óvulos que contém um ovário em estado de estro de um exemplar fêmea de Pirarucu de comprimento total médio de 1,90 m. é, aproximadamente, 180.505 óvulos em diferentes graus de desenvolvimento.
Esse total está distribuido, aproximadamente, da seguinte forma: a) óvulos maiores — 47.040; b) óvulos médios — 46.735; c) — óvulos pequenos — 86.730.
Portento, do total acima referido sómente 47.040 óvulos estao maduros, em condiçôes de serem lançados na agua para serem fecundados, na primeira desova, enquanto os outros continuam seu desenvolvimento com o crescimento normal e sucessivas mudanças de coloração.
Podemos verificar ser comparativamente pequeno o número de óvulos (47.040) que um exemplar fémea de Pirarucu de comprimento total de 1,90 m e 62 kg de peso dispôe para uma desova, quando uma Curunatá, Prochilodus argenteas, com 2,700 kg de peso pode concorrer com mais de 1.000.000 de óvulos para uma única desova (4) e um Dourado, Salminus maxillosus Cuvier, com 1 m. de comprimento total e 14 kg de péso, pode concorrer com mais de 2.000.000 (4). Todavia, em natureza, o aproveitamento final, em larvas, de cada desova é, sem dúvida, superior na primeira espécie, em virtude da proteçâo que os reprodutores dispensam aos ovos e larvas, fato que nâo sucede nas outras duas espécies.
Testiculos — O testículo dos exemplares machos adultos nessa espécie pode ser considerado, também, como orgão ímpar, pois em quatro exemplares que examinamos, de 1,80 m de comprimento total, médio, sómente encontramos desenvolvido o do lado esquerdo, enquanto o do lado direitó apresentava-se atrofiado, de comprimento bastante reduzido, não chegando a atingir 1/10 do outro.
O testículo esquerdo, no período de reprodução, quando apto a fornecer gàmetos normais, é alongado e semicilindrico, tendo a face plana voltada para baixo, livre, e a outra, a oposta, curva, fortemente aderida ao peritónío, em toda a sua extensâo, por um ligamento. Apresenta a parte mediana mais larga e de maior espessura. A parte cefálica é mais estreita e termina em ponta arredondada, enquanto a parte caudal é mais espessa. Um exemplar macho de 1,88 m de comprimento total ao ser necropsiado apresentou o testiculo esquerdo, o mais desenvolvido, medindo 260 mm de comprimento, enquanto o direito mediu apenas 26 mm.
HÁBITOS DE DESOVA
Sem dúvida, éste é um dos capitulos mais interessantes da biologia do Pirarucu. Multas lendas sobre os hábitos de desova dessa especie sao narradas por pessoas que visitaram a Amazonia. Por tais motivos, empreñamos todo esforço possivel no sentido de esclarecer referido assunto. Embora as nossas observaçôes tenham sido realizadas sóbre peixes mantidos em cativeiro, acreditamos, todavia, que em natureza os hábitos de desova não sejam multo diversos.
O período em que são observadas as desovas estende-se desde fins de Dezembro de um ano até fins de Maio do ano seguinte.
Inicialmente, quando ainda não haviam sido selecionados os casais de reprodutores, todos exemplares adultos eram mantidos em comum nos 12 tanques do viveiro. Os exemplares adultos não apresentavam caracteres sexuais secundários, senão poucos dias antes e depois da realização da desova.
— 403 —
observations were made on the fish in captivity, we believe that in nature spawning habits do not change very much.
The spawning period begins at the end of December of one year and ends by the end of March of the following year. At the start, when the couples had not yet selected themselves, all the adults were kept together in the twelve tanks. The adult fish did not show secondary sexual characteristies until a few days before or after spawning. It is true that we found out later that the females were longer and heavier than the males.
The first signs of spawning are indicated by fights among the couples for the possession of a special tank for their spawning. The fight is sometimes so tough between one couple and the others that some are forced to jump over the the benchlike partitions between the tanks, when chased by the others.
On the same day when the spawning signs are noted, one can already see the extra genital, secondary sexual characteristics in the reproducers. The male takes on a strong, dark coloration over the upper part of the head, and from there on to the dorsal region down to almost the insertion of the dorsal fin, while the sides, belly and tail regions take on a beautiful red coloration of the same color of the urucú seed (Bixa orellana). In the female, the change of coloration is less noted, all the fish surface taking on a light chestnut hue. The larvae are also black and swim over the head and back of the male that protect them. In captivity, when we noted that the fight for the possession of the tanks had subsided, we tried to help each couple by separating their tank from the others, by means of a gate made of strong wooden boards, as the wire-screening used in former days was easily torn by the big fish.
Among the conclusions arrived at after repeated observations on the spawning habits of this fish, we consider the question of depth of the spawning tanks as the most important. Thanks to the special conditions of the fish-ponds as far ae depth is concerned, the fish could enjoy water depths ranging from 0.80 to 1.90 m.
Out of the twenty spawnings observed, eleven took palce in shallow tanks, with depths from 0.80 to 1.00 m; five others in tanks of medium depths (1.30 to 1.50 m); and the other four in the deepest tanks (1.70 to 1.90 m). Therefore, it seems that the pirarucu prefers to spawn in, a place where the water depth does not exceed 1.00 m.
As regards the fertility of the species, only once had we the opportunity of collecting all the larvae for counting (on May 6th, 1948). The collected larvae were found to have a total length of 17.5 mm each and reached a yield of 11 403 fry. As for the sheltering of the spawning place, we found out that they prefer a spot as far removed from any vegetation as possible, be it floating or rooted vegetation. Out of the twenty spawnings, more than a half took place in tanks of even numbers, i, g., the ones stripped of any vegetation.
The pirarucu always endeavors to hide the effect that spawning has taken place, thus instinctively protecting the offspring. That is why it is very difficult for anyone to study the biology of this species from the egg stage.
In the Amazon Valley, the most absurd versions on the spawning habits of this fish go from mouth to mouth. It is a common story that the larvae suck milk from glands located in the reproducer’s head. Reference is also made of the possibility of reproducers hiding the eggs in their mouths during the incubation period and doing the same later on in regard to the larvae and small fry, to protect them from destruction by their natural enemies. Gudger (2) refers to this type of incubation in the pirarucu („oral gestation” or „buccal incubation”). However, as mentioned farther on, we did not observe this kind of incubation or protection to the larvae. The aforementioned author based his information on the fact that this incubation is done by two other kinds of fish that belong to the o Osteoglossidae family and on the information obtained on the almost exclusive method of catching this big fish with a harpoon (hooks are seldom used). Still according Gudger, when caught with hooks while fighting to get free, the fish releases the eggs which were incubating inside its mouth. (2).
The last of the reasons given above is weak, as in three out of six dams in Northeast Brazil, where the pirarucu was planted in 1940 by the Service de Piscicultura of Departamento Nacional de Obras Contra as Secas, after the reproduction capacity of this species was proved and its catching allowed (after 1947), more than 79 tons of fish were captured during only two months of fishing with the exclusive use of common hooks. And it should be noted that fish of both sexes with more than 2 m length were captured. Observing fish reared in tanks, we could better ascertain the above mentioned facts, although we still do not know some phases of the development of this species.
In almost all the spawnings that we observed since 1944, when the breeders were more than 5 years old, we could foresee the spawning by the struggle between reproducers. However, a period of complete calm would follow, in which nothing could be observed due to the water being agitated. From four to five days thereafter, we noted that the reproducers, each in turn, kept themselves in a vertical position, head downwards, tail showing above the water surface in the shallower tanks. After 3 to 5 days, the dark-backed reproducer kept its head lower in the water than the rest of its body. Only experienced people could see the larvae on this first day as they would never come to the surface. The school moves slowly in an ever-changing spheroidal formation close above the head of the dark reproducer — the male.
The reproducer directly responsible for the protection of the newly born larvae never leaves the school behind but is always before it, swimming slowly, rarely coming to the surface for breathing. Any shadow cast over the water by a less cautious person who comes too close makes the reproducers and the school descend to the bottom until they become entirely invisible.
The other reproducer — the female — keeps swimming at a short distance, also making circumferences, larger in radius, keeping off any other fish that tries to approach the school.
In twenty spawnings observed we never saw a reproducer protect the larvae by hiding them in its mouth.
When we unexpectedly came to the tank where the reproducer was looking after its school, it would neither try to leave the larvae nor hide them in its mouth. It would go down slowly, always keeping the shoal above its head, thus making it very easy for us to capture the larvae in their various stages of development with the help of a scap which had been equipped with a long handle.
The dark hue taken on by the head and back of the male works as a camouflage for the larvae which being also dark become almost invisible when swimming above the head of their keeper.
Only when they reach the length of 17.5 mm can the laevae be clearly seen. It is then that they start coming to the surface to breathe through their mouths. The school does not come to the surface in its totality, at one time, but rather in succession, at intervals varying from 60 to 85 seconds, during the first thirty days. After this period, as the small fish grow, the space of time between one such breathing and the next becomes longer. When the small fry become very lively, they jump for the needed air, with half their bodies coming out of the water.
More than once we took the opportunity — when the reproducer was protecting the larvae — to determine its sex.
As regards the knoweldge of the of the early phases of development of the pirarucu, we owe it to the opportunity we had on February 19, 1946. When the fish-ponds were being cleaned and the water renovated, for a better vision, there was a great water reduction. It was then that our attention was caught by the insistence that one dark-backed fish showed to remain at one determined spot of tank number two, near the water entrance, while another swam slowly at a certain distance. Due to the reduction in the water volume, all the other reproducers had moved on to deeper tanks. It was easy for us to make both reproducers move to deeper tanks and then examine the place where the dark-backed fish had been. We then found out a round cavity that on being explored by means of a small scap proved to contain together with silica clay humus of the bottom and small bits of water-weed roots, a great quantity of newly born larvae, eggs fully developed and bad eggs. The small hole (made in the natural earth of the tanks) measured 17 cm in diameter and 20 cm in depth. The edges showed some resistance and the bottom was roundish. Later, when cleaning the other tanks we discovered other such holes which differed only in diameter. In the ground which had a larger quantity of humus, the opening of the holes had a larger diameter which reached up to 50 cm, while depths were not more than 20 cm.
At the end of its evolution, the egg of the pirarucu is light green and both in color and shape looks very much like the quiabo seed (Hibiscus esculentus Linneu) when not yet ripe. It shows little resistance and has its roundish extremity
— 406 —
entirely opaque, while the other side shows some transparence. Its longitudinal diameter measures 4.2 mm and the transversal one 2.8 mm.
The newly born larva has the upper part, from head to tall, transparent and the lower part, the one which corresponds to the yolk sac completely light green.
This voluminous food reserve of the pirarucu larvae differs from those of other fish, because it is elongated, with 7.5 mm in length and 1.9 mm in width, narrowing towards the tail region. The total length of the larva is 11.6 mm when it comes out of the egg and its heart beats 94 times per minute.
Twenty-four hours af ter the larva comes out of the egg it measures 13,5 mm in its total length, the eyes already pigmented. There are 112 heart beatings, sparse pigments appear on the transparent upper part (considered the very body of the larvae) and the yolk sac shows a lighter green coloration.
In its third day, the larva is 14.6 mm long, 130 cardiac beating are registered, there is more pigmentation on the body, the gill filaments are clearly seen, and the eyes, 0.4 mm in diameter, are more strongly pigmented.
The larvae are still in the same position as when they were born. They remain lying on their sides due to the heaviness of the yolk sac that is being absorbed. However, they do not keep still. There is an oscillating movement in the body less accentuated from the tail to the head.
On the fourth day, the larva reaches 15 mm in its total length, 158 heart beatings are registered, the head is well pigmented, the eye diameter is 5 mm, the mouth and anal orifice are already open although external food has not yet been used. The yolk sac, in large part already absorbed, is reduced to 6.4 mm in total length and the pectoral fins can already be seen.
At the beginning of the fifth day, helped by the pectoral fins, and having the yolk sac greatly reduced, the larvae, now 15.8 mm long, started to lift themselves above the tank bottom. On the sixth day, the larvae were swimming easily, although the yolk sac had not yet been entirely absorbed.
The gills are partly protected by their covers. On the larva body, three deeply pigmented stripes can be noticed: two along the sides, running from the humeral region until near the anus and one on the upper part, going from the middle of the head to the beginning of the dorsal insertion, in a narrowing progression, as in the first two stripes.
In view of the absence of the male that protects them, the larvae swim aimlessly, scattered, close to the water surface. For the first time, micro-crustacean eggs and algae were found in some larvae submitted to necropsy, with a total length of 16,5 mm, and having the digestive tube already open.
On their seventh day of free life, almost ali the larvae measured 17.1 mm and their gills were almost entirely protected by their covers. The dark pigmentation characterized them by this time. They were still swimming aimlessly, dose to the surface, open-mouthed; as they were trying to catch some food. The yolk sac is now further reduced, the cephalic portion showing dark yellow, and the end portion being light yellow. In all larvae examined the digestive tube contained micro-crustacean and algae.
— 407 —
On the eighth day, all the larvae had absorbed almost entirely their food reserve (yolk sac), were now 17,9 mm long, and their disgestive tube, and their disgestive tube were full of micro-crustacea and algae.
On their ninth day of free living, for the first time we saw a school of larvae swimming in formation, ascending from moment to moment to the water surface for breathing, as if they were under the protection of the reproducers. Their total length then was 18.5 mm each.
After these observations, we concluded that the pirarucu larvae, contrary to what is commonly heard in the Amazon Valley, can be reared like any other fish larvae, i. e., artificially, by the thousands, and altogether separated from the reproducers, provided that at a convenient time, after their fifth day of outside life, they are given nanoplankton rations (algae and micro-crustacean). This epoch coincides with the permeability of the digestive tube, when the food supply in the yolk sac is about to be completely absorbed.
Reared under the protection of the reproducers, the larvae can only be perfectly noted after the seventh day, when they measure more than 17 mm of total length. They then come to the surface and start to get food in their surroundings.
Under the reproducer’s protection, the larvae always swim together, in a sole group, what does not happen when they are artificially reared, separated from their parents, since it is then noticed that they group themselves in several schools, although they sometime get together forming one sole group.
When they reach 40 mm in total length, the young fish are transfered to the rearing tanks and are fed plankton mixed with ground fish meat.
After 60 days of their hatching, the young fish artificially reared reach the total length of 100 mm and have the following characteristics and measurements: pectoral fin — length: 8.5 mm, number of rays: 11; belly fin: length 4.8 mm; number of rays: 6; anal fin: 23.7 mm long, 31 rays; tail fin: 10 mm long, 28 rays; dorsal fin: 23.1 mm long, 34 rays; length of head: 28.7 mm; diameter of eye: 3.9 mm; total length of intestines, from the pylorus to the anus: 78.8 mm; two pyloric appendixes measuring 17.4 mm and 11.2 mm respectively; stomach in V form, of uneven branches, measuring 15,5 mm and 12.3 mm respectively.
When they reach this length, the young fish already eat shrimps and small piabas (Tetragonopterinae) cut into small pieces.
The artificial rearing of the pirarucu is relatively easy. It is an essentially carnivorous fish. The reduced length of its stomach leaves no doubt about it. In necropsies made in 1 m 90 long pirarucus, we found out that their intestines measured 2.45 m from the pylorus to the anus.
During the rearing of the young fish, when it is hard to find fish small enough to feed them, the best solution is to cut the food into small pieces.
— 408 —
R E S U M E
Having conducted observations and research on various adult pirarucus, Arapaima gigas (Cuvier), kept in fish-ponds at the Posto de Piscicultura of Lima Campos, Icó, Ceará Brasil, the author came to the following conclusions:
a) each fish of this species has a sole gonad;
b) reared in captivity they reach sexual maturity after their fifth year;
c) during the reproduction period,there appears in the males an extragenital, secondary sexual characteristic, namely, the darkening of head and dorsal regions in addition to a change in coloration which becomes red on the scale edges of certain regions of the fish’s body;
d) the species is classified in the group of fish which have partial sexual maturation:
e) the eggs are spheroidal in form, light green, with transversal and longitudinal diameters of 4.2 mm and 2.8 mm respectively;
f) the male reproducer is directly responsible for the protection of the school of larvae and small fry;
g) contrary to what was believed no oral incubation was observed;
h) reproducers were never seen to protect the larvae of young fish by hiding them in their mouth;
i) there is no lactation phase in the life cycle of the larvae;
j) soon after hatched, the larvae can be artificialy reared;
k) soon after hatched, the larvae measure 11.6 mm in length;
l) the yolk sac of the larvae is large and elongated;
m) the larvae start to look for food from the outside world after their fifth day of free life.
— 409 —
1. EIGENMANN, C. H. & ALLEN, W. B., 1942, Fishes of Western South America, University of Kentucky, Lexington, Ky., U. S. A., pp. 338-345, 6 figs.
2. GUDGER, E. W., The giant fresh-water fishes of Sout America. Scl. Monthly, 57:500-513.
3. IHERING, R., von, 1907, Os peixes da agua doce do Brasil. I. Gymnoti (Peixes espada, Tuvira, etc.). Cichlidae (Acará, papa-terra, etc.) Rev. Mus. Paul., S. Paulo, 7:258-336, 7 figs., I est.
4. IHERING, R, von, & Azevedo, P., 1943, A Curimatá dos açudes Nordestinos, Arp. Inst. Biol., S. Paulo, 5:143-184, ests. 4-9.
5. MYERS, G. S., 1943, Sistemática Gerai de Peixes e Biologia da Pesca. Apontamentos de Curso. Museu Nacional, Rio de Janeiro, VIII -|- 84 pp. mimiogr. (Rev. Brasil. Biol., 8 (4) : 445-459 — Dezembro, 1948 — Rio de Janeiro, D. F.)