You are viewing the archived website of Pond Dynamics / Aquaculture CRSP. When using this website, please understand that links may be broken and content may be out of date. You can view more information on the continuation of PD/A CRSP research archived at AquaFish Innovation Lab.
9NS3-Development of Sustainable Pond Aquaculture Practices for Colossoma macropomum in the Peruvian Amazon
PD/A CRSP Eighteenth Annual Technical Report
Previous Section
Table of Contents
Next Section

Cite as: [Author(s), 2001. Title.] In: A. Gupta, K. McElwee, D. Burke, J. Burright, X. Cummings, and H. Egna (Editors), Eighteenth Annual Technical Report. Pond Dynamics/Aquaculture CRSP, Oregon State University, Corvallis, Oregon, [pp. ___.]

Development of Sustainable Pond Aquaculture Practices for
Colossoma macropomum in the Peruvian Amazon

Ninth Work Plan, New Aquaculture Systems/New Species Research 3 (9NS3)
Progress Report

Christopher C. Kohler
Fisheries and Illinois Aquaculture Center
Southern Illinois University at Carbondale
Carbondale, Illinois, USA

Susan T. Kohler
Economic and Regional Development Office
Southern Illinois University at Carbondale
Carbondale, Illinois, USA

Marcos J. De Jesus
Fisheries and Illinois Aquaculture Center
Southern Illinois University at Carbondale
Carbondale, Illinois, USA

Fernando Alcántara and Salvador Tello
Instituto de Investigaciones de la Amazonia Peruana
Iquitos, Peru

Printed as submitted


Colossoma macropomum growth performance did not significantly differ in trials conducted in ponds at 2,500, 3,250, and 4,000 fish ha-1 in Iquitos, Peru. Fish initially weighing 3.4 g were fed a locally prepared diet (26.7% crude protein; 9.0% crude lipid) in rations ranging from 3 to 5% body weight per day. Fish were harvested after 168 days and had mean weights of 374.7, 307.7, and 287.0 g for the 2,500, 3,250, and 4,000 fish ha-1 stocking rates, respectively. Survival ranged from 67 to 96%, though all but two of nine ponds exceeded 80% survival. Feed conversion efficiency was 40.4, 43.4, and 61.3%, respectively, for the 2,500, 3,250, and 4,000 fish ha-1 treatments. Fish in two of the ponds were reared for an additional five months and attained a mean weight of 1 kg. Water quality parameters remained within acceptable ranges for tropical aquaculture. As with Piaractus brachypomus in a previous study, this study suggests the economic feasibility of rearing Colossoma in the Peruvian Amazon. Generally, the combined cost of fingerlings (US$0.14 each; corrected for 90% survival) and feed (US$1.02 kg-1 to produce 1 kg fresh fish) is under half the price (US$3.00 to $4.00 kg-1) for which the fish are sold in the Iquitos market during flood periods.


Native species aquaculture has been expanding in the Peruvian Amazon as research has played a major role in the positive evaluation of its potential. Alike Piaractus brachypomus, Colossoma macropomum is native to the Amazon basin, and shares many characteristics that also makes it suitable for aquaculture. Local production of this species is still in an extensive manner, but it possesses a high demand and attains a higher price at the market. This motivates local farmers to invest their time in the production of this valued fish. As research develops, more important information becomes available to producers, hence the improvement of native species aquaculture in the region. However, technology is still under-developed, but external aid has made it possible for the locals to become more aware and active with aquaculture practices of Colossoma and other native species.

Colossoma broodstock are many times collected from their habitat although they are also raised in captivity at research stations. These fish may be immediate descendants of wild broodstock or may even be a product of multiple generation breeding. Normally a large broodstock population is reared for wide selectivity and to increase the probability of successful reproduction. They are then selected upon external physical characteristics when ready to spawn.

No standardization exists for stocking densities for fry or fingerlings (Campos, 1993). Likewise, no uniform fish diets are available in the region (Cantelmo et al., 1986; Ferraz de Lima and Castagnolli, 1989). The purpose of this study is to determine suitable stocking densities for optimal and efficient production of Colossoma macropomum to market size (0.5 to 1.0kg) using a prepared diet, manufactured from locally available ingredients. Replicated pond studies were carried out at the Instituto de Investigaciones de la Amazonia Peruana (IIAP) research facilities at Iquitos.

Along with this density study, other experiments related to the performance of these fishes are currently in progress. A nutritional study performed by Rebecca Lochmann was set to determine an optimal broodstock diet at low cost, manufactured with locally available ingredients. Konrad Dabrowski is completing an experiment where he studies the fish's blood plasma steroid concentration during their annual breeding cycle. A study to compare the effects of different inducing hormones on Colossoma and/or Piaractus is also in progress. Asecond density study with Piaractus is presently under completion at the host country research facility. All together these studies will provide important information to develop an efficient protocol for the spawning and rearing of the Amazon fish species.

Methods and Materials

Nine ponds, ranging in size from 1,015 to 5,320 m2, were stocked with Colossoma macropomum at three densities: three ponds at 2,500 fish ha-1, three at 3,250 fish ha-1, and three ponds at 4,000 fish ha-1. The mean initial weight was 3.4 g. A locally manufactured feed using available ingredients was fed (Table 1). Fish were fed 5% body weight per day (bwd) for the first month and 3% bwd for the remainder of the trial. Rations were divided into three feedings. Fish were sampled (10% minimum population) by seining every two weeks to record lengths and weights. At harvest, biomass, feed conversion efficiency (FCE), specific growth rate (SGR), and condition factor (K) were calculated. The study commenced 17 April 1997 and continued until 20 October 1997. General water quality parameters (dissolved oxygen, temperature, total ammonia nitrogen, and pH) were measured daily or weekly, in the early morning. Harvest data were analyzed using the Statistical Analysis System (SAS Institute 1993) with an alpha of 0.05.

Materials and methods, as well as results for the nutrition and blood plasma studies will be presented respectively by Rebecca Lochmann and Konrad Dabrowski in their reports. The remaining density study and reproductive hormone experiment will be reported in the final workplan report.


No differences (P > 0.05) existed at harvest in C. macropomum weight (374.67 g at 2,500 fish ha-1, 293.5 g at 3,250 fish ha-1, and 377.33 g at 4,000 fish ha-1; see figure 1), total length (26.03 cm at 2,500 fish ha-1, 24.75 cm at 3,250 fish ha-1, and 26.33 cm at 4,000 fish ha-1), specific growth rate (2.32 at 2,500 fish ha-1, 2.60 at 3,250 fish ha-1, and 2.61 at 4,000 fish ha-1), condition (2.1 at 2,500 fish ha-1, 1.9 at 3,250 fish ha-1, and 2.1 at 4,000 fish ha-1), feed conversion efficiency (40.4 at 2,500 fish ha-1, 43.4 at 3,250 fish ha-1, and 61.3 at 4,000 fish ha-1), or productivity (894.9 kg ha-1, 715.3 kg ha-1, and 1098.3 kg ha-1; see table 2). Survival exceeded 80 %.

Water quality varied among ponds (Table 3). Mean maximum and minimum temperatures over the course of the study were 31 and 28.5 C, respectively. Minimum dissolved oxygen levels generally remained above 1.0 mg L-1, and usually averaged in excess of 4.0 mg L-1. Total ammonia nitrogen remained below 1.0 mg L-1. Carbon dioxide was not recorded this study, but levels reached a high of 22 mg L-1 in one pond during the last trial. These waters can be classified as soft (hardness = 20 mg L-1; alkalinity = 20 mg L-1; conductivity = 96 ohms cm-2) and slightly acidic (morning pH ranging from 3.4 to 9.1).


No significant differences were found in growout performance of Colossoma macropomum stocked in ponds at densities of 2,500, 3,250, and 4,000 fish ha-1. The mean fish growth rate of 1.90g d-1 was lower than Piaractus in the previous study. Although, this can be attributed to certain ponds emerging with significant macrophyte infestations, which impeded the feeding process for an undetermined period of time. Some of the clearer ponds showed growth rates at about 3.0g d-1, which is what is expected of this species according to studies by St. Paul 1986 and Gunther and Boza Abarca 1992.

Feed conversion was excellent throughout the trial. Exceptionally high values indicate the ability of Colossoma to filterfeed. In contrast to Piaractus, they are equipped with longer, finer gillrakers. In both species, filterfeeding is evident. Also important, is the fact that various seeds and fruit that grow around the banks of the ponds are ingested by the fish when these fall in the water. Fish that were fed for an additional five months (10 months total) reached about a kg in size (from 3.4g). The prepared diet used in this study cost U.S. $1.02 to produce 1 kg of whole fish. Fingerlings generally sell about U.S. $0.13 each. Colossoma will attain a market price of over U.S. $3.00 kg-1.

Water quality remained well within the tolerances of Colossoma throughout the study. Truly, it must be recognized that this species has the ability to slow its metabolic rate during periods of stress. In fact, they can generate extended tubercles from their lower tip of the jaw that help them breath oxygen from the surface. So it would take days of low oxygen levels to adversely affect this species instead of hours, which is the case in many culture species.

This study revealed considerable potential for intensive aquaculture of Colossoma in the Peruvian Amazon. No significant differences were found between the three densities. Densities of 2,000 to 3,000 fish ha-1 are traditionally used in the region. According to results in this study, higher stocking densities may be possible. On 2 March 2000, triplicated ponds were stocked at 4,000, 6,000, and 8,000 P. brachypomus ha-1 at the same facility. This study is in its final stages, and harvest is coming up close. More comparisons will be available further on.

Anticipated Benefits

Most farmers generally use organic fertilizers and periodically provide fruits, nuts and kitchen scraps. This research presents an economic prepared diet (26.7% protein and 9% lipids) used for the growout of Colossoma. Considering the excellent growth rates that occurred (3.4g to 1 kg in ten months), it appears that this diet meets or exceeds Colossoma's nutritional needs. More detailed nutritional studies performed by Rebecca Lochmann should recommend a well balanced diet for Colossoma and Piaractus that could be manufactured locally with locally available ingredients at a low cost. Results of the present and ongoing studies will be shared with local farmers via extension work.

Konrad Dabrowski will give us a perspective as to when these fish reach their peak during their annual reproductive cycle in order to better program spawning efforts. Along these measures, he will also investigate how to obtain best gamete quality from the broodstock. Compiling all this valuable information will allow us to efficiently and successfully culture this important fish in the Peruvian Amazon.

Literature Cited

Campos, L.B., 1993. The culture of gamitana (Colossoma macropomum, Cuvier, 1818) in Latin America. M.S. thesis. Southern Illinois University, Carbondale, IL 148 pp.

Cantelmo, A., A. De Soura, and J. Senhorini, 1986. Dimencao da particula do alimento para alevinos de pacu, Colossoma mitrei e tambaqui, Colossoma macropomum. Proyecto Aquicultura (Editor), Sintese Dos Trabalhos Realizados Com Species Do Genero Colossoma, March 1982, Abril 1986., Pirassununga, Brasil, 28 pp.

Ferraz de Lima, J. and N. Castagnolli, 1989. Reproducao, larvicultura e genetica: cultivo de Colossoma. A. Hernandez (Editor), Primera Reunion Grupo de Trabajo Tecnico, Junio 1988. Pirassununga, Brasil, pp. 315-322.

Gunther, J. and J. Boza Abarca, 1992. Growth performance of Colossoma macropomum (Cuvier) juveniles at different feed rations. Aquaculture and Fisheries Management 23:81-93.

Saint-Paul, U., 1986. Potential for aquaculture of South American freshwater fishes: a review. Aquaculture 54:205-240.

Previous Section
Table of Contents
Next Section