-PROGRESS 
REPORT 
ON
FISHERIES DEVELOPMENT 
IN
NORTHEAST BRAZIL
July 1, 1975- December 
31. 1976
: .. .. : .... . .....:
July 1977
Project: Contract AID/csd - 1152, Task Order 2
International Center for Aquaculture
Research and Development Series No. 14
Agricultural Experiment Station, R. Dennis Rouse, Director
Auburn University, Auburn, Alabama
I- I
I
CONTENTS
Page
INTRODUCTION ............ ............... .................... 
3
TRAINING AND TECHNICAL ASSISTANCE ............................... 
3
INSTALLATIONS AND FACILITIES .............. ........................ 
4
AQUACULTURAL RESEARCH PROGRAM ................................ 4
Experim ents with Tilapias ................................... 4
Experiments with Tilapia and Tucunare ....................... 6
Experiments with Pirapitinga (Colossoma bidens) ............... 7
Experiments with Tambaqui (Colossoma macropomum) ......... 8
Experiments with Curimata Comum (Prochilodus cearensis) ...... 8
Experiments with Curimata Pacu (Prochilodus argenteus) ........ 9
Experiments with Channel Catfish (Ictalurus punctatus) .......... 9
FISH CULTURE EXTENSION............................................... 9
BIBLIOGRAPHY ........................................................ 11
PUBLISHED 6/77 - IM
Information contained herein is available to all without regard to race,
color, or national origin.
PROGRESS REPORT ON FISHERIES
DEVELOPMENT IN NORTHEAST BRAZIL
July 1, 1975 - December 31, 1976
Leonard L. Lovshin*
INTRODUCTION
AUBURN UNIVERSITY has been involved in improving the
freshwater fisheries resources in Northeast Brazil since 1966
when various staff members participated in short-term work
programs in that country. Since 1969, the University has
provided technical assistance in the form of resident advisors
to the Brazilian federal agency, DNOCS', in fish culture,
reservoir management, and fish culture extension. The
author arrived in Fortaleza, Ceara, in June 1972 and has
been the fish culture advisory specialist since that date.
This report deals with the pertinent activities related to fish
culture within the Centro de Pesquisas Ictiologicas during the
period July 1, 1975 - December 31, 1976. Past developments
in fisheries and fish culture in the Northeast of Brazil can be
found in previous progress reports (1, 2, 3, 5, 8).
* Assistant Professor, Department of Fisheries and Allied Aqua-
cultures.
Departamento Nacional de Obras Contra as Secas.
The topography of Northeast
dairy and calf feedlots.
TRAINING AND TECHNICAL ASSISTANCE
The Centro de Pesquisas Ictiologicas is rapidly becoming
recognized as one of the outstanding freshwater research
organizations in Latin America. Demands on the research
center's staff are continually increasing as requests for techni-
cal assistance and training increase. In the past year and a
half, DNOCS biologists gave technical assistance to six
Brazilian agencies located in the states of Bahia, Minas
Gerais, Goias, Acre, and the Federal District. Assistance
consisted of orientation and planning of fishery and
limnological surveys and aid in selecting sites for fish
hatcheries.
At the request of the Peruvian government, two DNOCS
biologists and the author spent a month in Peru touring
fisheries installations and advising Peruvian biologists on the
best methods of culturing selected species of Amazonian
fishes. During the reporting period, 50 students and scientists
,spent short study periods at the Centro de Pesquisas Ictio-
logicas, including 2 from Peru, 2 from Chile, 1 each from
Brazil is ideal for construction of inexpensive fish ponds. The ponds shown are fertilized by wastes from adjacent
4
Panama and Colombia, 
and 10 student trainees from 
local
universities. The local university 
students were in their final
year of studies and had won scholarships 
to study for a period
of 6 months to 1 year in the discipline 
of their choice.
DNOCS also administered 
a third course for 11 SUDEPE
2
scholarship students. These 
students spent 2 months 
at
DNOCS fishers installations, 
including 1 month with the
Centro de Pesquisas Ictiologicas. 
Upon completion of this
special training, most of 
the graduates departed for 
their
respective jobs with SUDEPE.
Recently, four officials 
from the Colombian government
visited DNOCS fishery installations 
to personally evaluate
the research center's potential 
as a fisheries training center.
These fisheries officials were 
interested in short term training
for Colombian biologists in 
the practical aspects of fishery
biology and fish culture.
Short term training is now 
available through DNOCS in
such specialty areas as 
intensive and extensive 
fish culture,
reservoir management, collection 
of commercial fishery
statistics, hatchery management 
including the spawning and
culturing of a number of 
Amazonian fishes, hormone
spawning of fish, and reservoir 
limnology.
INSTALLATIONS AND FACILITIES
The headquarters unit for the 
Centro de Pesquisas Ictiolo-
gicas presently is located in Fortaleza, 
Ceara. It is anticipated
that the entire Centro dle Pesquisas 
Ictiologicas will move to
Pentecoste, Ceara, when permanent 
facilities are completed.
DNOCS has embarked on 
a large-scale building program
in Pentecoste located 100 
kilometers west of Fortaleza.
Programmed for 1976 and 
1977 are the construction of 
a
1,000-square meter 
laboratory and teaching 
complex,
administration building, 
warehouse, garage-machine 
shop,
fish technology building, and 
net fabrication and storage
house. The building complex 
will be surrounded by pond
facilities on 120 hectares 
of land located below the 
Pereira de
Miranda Reseroir at Pentecoste. 
A restaurant, dormitorn,
and housing for technicians 
and visiting scientists are
planned for construction 
during the second phase 
of the
building program. Construction 
has been completed recently
on ten 2,500-square meter 
earthen ponds; a small 
fish
hatchery unit modeled after 
larger DNOCS hatcheries that
SSuperintendencia do Desenvolvimento 
da Pesca.
DNOCS has made a commitment to 
fishculture research by developing
a research facility which presently contains 151 
earthern and concrete
ponds with a total water surface of 11.0 hectares. The 0.4-hectare 
ex-
perimental units shown are typical 
of the excellent research ponds
available.
contains 28 concrete ponds ranging 
in size from 40.3 to
80.5 square meters; six 100-square meter 
earthen ponds for
tilapia hybrid spawning experiments; 
and a wet-laboratory
unit that will house twenty 500-1 cement-fiber 
tanks and
forty 20-1 aquaria. At present, the Centro de Pesquisas 
Icti-
ologicas has 151 earthen and concrete 
ponds ranging in size
from 40.3 to 10,000 square meters. 
Total water area is 11.0
hectares.
When the research center is completed 
in 1980, the facility
will contain over 200 earthen 
ponds, comprising over 20
hectares of water, and related 
wet-laboratory installations.
DNOCS will possess excellent 
installations for research and
training in fish culture, 
reservoir management, 
and
limnology. With the help 
of USAID, DNOCS hopes 
to
develop Pentecoste installations 
into a regional center for
fisheries training and research for 
Latin America.
AQUACULTURAL RESEARCH PROGRAM
A program of aquacultural research has 
been in progress
since 1970, directed towards evaluating Brazilian fishes for
culture potential and developing culture systems 
for use in
DNOCS irrigation projects and by private 
farmers. Exotic
species with known culture potential are also being 
investi-
gated. This report summarizes research performed over 
the
past year and a half, primarily in the earthen pond research
complex.
Experiments with Tilapias3
At present, tilapia offer the greatest potential 
for
immediate culture in Northeast Brazil. Past 
research results
can be reviewed in publications 
by Lovshin, Da Silva, and
Fernandes (7), Lovshin (8), and Lovshin and Da 
Silva (9). The
majority of research on tilapias is presently centered 
on the
all-male tilapia hvbrid 
produced by crossing 
male
Sarotherodon hornorum with female Sarotherodon 
nilotica.
I ilapia species reported on in this paper haoe been 
reclassified
recentl\ and presentlk are included in the gemnus 
Sarotherodon.
I I
The tilapia hybrid is an excellent culture 
fish and enjoys wide consumer
acceptance. Production of the tilapia hybrid 
when stocked at 30,000
fingerlings per hectare and fed agricultural 
by-products reaches 10,000
kilograms per hectare per year 
with an average size of 400 grams.
However, widespread, large scale 
culture is limited by short supplies
of fingerlings. Research now being 
conducted may provide hatchery
methodology to alleviate this problem.
Tilapia hybrids continue to give excellent results 
when
intensively cultured utilizing agricultural 
waste products or
organic manures as feed and fertilizer. 
Past research
demonstrated that tilapia hybrids stocked 
at 10,000 per
hectare and fed agricultural waste products 
reached an
average weight of 200-300 grams in 6 months, producing
2,000 to 3,000 kilograms per hectare.
To test the effect of increased 
stocking rates on production,
DNOCS biologists investigated the growth and production 
of
tilapia hybrids when stocked at 
13,000, 15,000, 17,000,
19,000, 21,000, 23,000, 25,000, 27,000, 29,000, and 
31,000
per hectare. The fish were stocked in 350-square 
meter
earthen ponds and each stocking rate was represented 
by a
single pond.
The hybrids were fed a ration of 50 percent babacu 
cake (a
type of palm nut cake containing 24 percent protein) 
and 50
percent cottonseed cake (23 percent protein) at 5 percent of
the total weight of the fish in each pond. The fish were 
fed 5
days a week. Feeding rates were recalculated each month
based on seine samples of the ponds. Sampling consisted of
weighing and measuring 50 fish from each pond. Each 
pond
was fertilized with a total of 19 kilograms (540 
kilograms per
hectare) of triple superphosphate and ammonium sulfate
applied at 2-week intervals over the first 7 months of 
the
experiment. Table 1 shows the results of this experiment 
at
the end of 367 days.
Productions were the highest vet obtained in research with
the tilapia hybrid. The productions are even more impressive
considering that only vegetable materials were fed to the fish.
It is the first time that 1 kilogram of marketable size 
fish has
been obtained from 1 square meter of standing water at
Pentecoste. Preliminary analysis of the data indicated that
increasing stocking rates had a positive effect on production.
Conversion of feed to fish flesh was good 
up to 6 months,
average at 10 months, and poor at 12 months. 
From the
tenth month until harvest, conversion 
rates increased sharply,
probably caused by extremely poor water quality.
The purpose of the experiment was to determine at what
time growth rates decreased and mortality occurred with 
the
tilapia hybrids when stocked in standing water and 
fed
agricultural by-products. For this reason, fish were 
fed and
the ration was increased even when fish were obviously 
being
stressed by poor water quality. The tilapia hybrid 
is
extremely tolerant. Fish were observed at the water 
surface
I 55
night after night seeking oxy gen during the last month of 
the
experiment. A 24-hour dissolved oxygen test 
was run 1 week
before harvest. Dissolved oxygen readings 
were taken at the
water surface and near the bottom in all 
ponds at 4-hour
intervals starting at 12:00 noon. Tilapia 
hybrids survived in
water having less that 0.8 p.p.m. of dissolved 
oxygen for 8 to
10 hours (luring the night. Four days after 
the extremely low
dissolved oxygen readings were recorded, 
the first fish kill of
tilapia hybrids at Pentecoste occurred. 
It is not known if the
kill was a direct result of low dissolved 
oxygen or a
combination of several water quality factors. 
Several other
TABLE 1. SL\MMtsAM OF RESULTS \WITH TIL\PI .A HBHIDS STOCKED AT VARIOUSLEVELS
Performance measure 
Result, by stocking rate per hectare
13,000 15,000 17,000 19,000 21,000 23,000 
25,000 27,000 29,000 31,000
Fish stocked/pond........... 455 525 
595 665 735 805 875 
945 1,015 1,085
Av. initial weight, g ......... 25 27 
26 38 22 22 23 
23 25 25
Av. final weight, g .......... 452 466 
495 456 492 384 444 
416 353 410
Production pond, kg ........ 197 245 
279 303 354 309' 371 
375 356 419
Production ha, kg ........... 5,629 7,000 
7,971 8,657 10,114 8,828 10,599 
10,714 10,171 11,971
Survival, pct................ . 96 100 
95 100 98 100 96 
95 99 94
Feed conversion at 6 months.. 3.3 2.8 
2.7 3.2 2.4 2.8 3.3 
3.0 3.0 2.7
Feed conversion at 10 months. 4.5 4.6 
4.0 4.5 4.4 4.1 3.8 
4.2 4.3 3.7
Feed conversion at 12 months. 6.0 6.3 
6.2 6.2 6.2 6.7 5.9 
6.2 6.3 6.2
Ration fed/pond, kg ......... 1,124 1,465 
1,648 1,732 2,093 1,947 2,074 
2,168 2,067 2,413
SIncluded are 62 kg of fish that died due to poor water quality 2 
days before harvest.
Productions of 2,900 kilograms per hectare per year can be obtained by
washing hog manure and feed wastes from pens into ponds stocked
with 8,000 tilapia hybrids per hectare. Fish average 206 grams after 6
months of culture.
ponds being fed at a higher daily rate suffered no fish
mortality. The pond stocked at 31,000 per hectare was fed
daily 20 kilograms (564 kilograms per hectare) for 2 months
without a fish kill.
These data indicate that even higher productions can be
expected from tilapia hybrids under the experimental
conditions present at Pentecoste.
Da Silva, Lovshin, and de Melo (unpublished data) tested
the all-male tilapia hybrids raised in conjunction with feeder
pigs. Three ponds of 1,000 square meters were stocked with
25-gram hybrid fingerlings at the rate of 8,000 per hectare.
Pig sties were constructed on the margins of three ponds and
7 pigs (70 per hectare) averaging 17 kilograms were placed in
each sW
. 
The pigs were not allowed to enter the ponds. The
sties were cleaned daily and all waste products washed into
the ponds. The pigs were fed daily at 5 percent of their body
weight a ration consisting of manioc-30 percent, wheat bran-
20 percent, corn-15 percent, babacu cake-15 percent, and
grass-20 percent. This ration contained 10 percent protein.
After 189 days, 1,490 kilograms per hectare of tilapia
hybrids averaging 205 grams each were harvested. Pigs
averaged 60 kilograms each. Conversion rate of feed to pigs
was 7.1 to 1; of feed to pigs and fish combined, the rate was
5.9 to 1.
Da Silva, Carneiro-Sobrinho, and de Melo (unpublished
data) recently terminated a second experiment similar to the
above mentioned one, testing tilapia hybrid and pig culture.
Six pigs per sty (60 per hectare) were fed a daily ration of
corn, babacu cake, meat meal, and grass at 5 percent of their
body weight. The ration contained 11.6 percent protein.
Tilapia hybrids averaging 31 grams each were stocked at the
rate of 10,000 per hectare. The fish were fed cottonseed meal
at 2 percent of their body weight, 6 days a week. After 193
days, 3,043 kilograms per hectare of tilapia hybrids
averaging 304 grams each were harvested. Average initial
weight of the pigs was 16 kilograms and average final weight
was 55 kilograms. Conversion rate of feed to pigs was 6.4 to
1 and of feed to fish it was 1.7 to 1.
Three ponds of 1,000 square meters were also stocked with
25-gram tilapia hybrid fingerlings at the rate of 8,000 per
hectare and fertilized with chicken manure at 50 kilograms
per week (500 kilograms per hectare per week). The chicken
manure contained 16 percent protein and consisted of 79
percent organic matter. After 189 davs, 1,350 kilograms per
ectare of tilapia hybrids averaging' 186 grams each were
harvested. Conversion rate of chicken manure to fish was 10
to 1.
Experiments with Tilapia and Tucunare
Because of the problems encountered in producing large
numbers of tilapia hybrid fingerlings, DNOCS biologists are
now studying the production capacity of Sarotherodon
nilotica when stocked with a predator fish species. Besides the
biological problems involved with producing and
maintaining pure strains of broodstocks, DNOCS fish
hatcheries do not have sufficient facilities to produce the
large numbers of tilapia hybrid fingerlings needed to meet
present demand. However, with existing installations, large
numbers of pure Sarotherodon nilotica fingerlings can be
easily produced. The predator tucunare, Cichla ocellaris, has
6
A'
~i* ~
been produced for many years in DNOCS hatcheries and
production of large numbers of fingerlings of this species is
easily accomplished.
To test the efficiency of tucunare in controlling the
excessive reproduction of tilapia, various ratios of tucunare
and Sarotherodon nilotica were stocked into 400-square
meter earthen ponds. These ponds were new and exhibited a
higher than normal rate of seepage. All six ponds were
stocked with pure Sarotherodon nilotica fingerlings at the
rate of 10,000 per hectare. Predators were stocked in the
predator to tilapia proportions of 1:3, 1:6, 1:9, 1:12, and
1:15. One control pond did not receive tucunare.
All ponds received a total of 1,349 kilograms (33,725
kilograms per hectare) of cattle manure applied at
approximately 2-week intervals. Fish were fed available
agricultural waste products, which included rice and wheat
bran, castor bean meal, and babacu cake at 5 percent of their
body weight for the first 6 months and 3 percent of body
weight the remaining 4 months. The fish received the ration 6
days a week. Feeding rates were changed monthly using the
results of monthly seine samples.
After 302 days, ponds were drained and all fish were
counted and weighed by species, sex, and size group. An
average of 2,491 kilograms per hectare was produced with a
predator while the control pond without a predator produced
4,944 kilograms of tilapia per hectare, or about twice the
yield with a predator, Table 2. However, few tilapia reached
marketable size in the control pond. Total control of tilapia
reproduction was not obtained in any pond, but only in the
pond with a ratio of 15 tilapia to 1 tucunare did fewer than
90 percent of the fish reach marketable size. While male
tilapias grew twice as fast as females, all the stocked females
reached harvestable size. Survival of both Sarotherodon
nilotica and tucunare was poor. The reason for this is not
known. This initial experiment demonstrated that good
production of Sarotherodon nilotica can be obtained with
tucunare as a predator to control reproduction at ratios of up
to 12 tilapia to 1 tucunare. The predator-prey ratio of 15
tilapia to 1 tucunare gave inconclusive results because of low
predator survival and will have to be retested.
TABLE 2. SUMMARY OF PERFORMANCE OF SAROTHERODON NILOTICA RAISED
IN ASSOCIATION WITH THE PREDATOR, CICHLA OCELLARIS,
AT VARIOUS RATIOS
Result, by ratio of S. nilotica to predator
Performance All
measure tilapia 8:1 6:1 9:1 12:1 15:1
Av. initial wt. of
predator, g ...... . ---- 12 11 
11 10 13
Av. initial wt. of
S. nilotica, g ..... 
14 15 14 
15 14 14
Av. final wt. of
predator, g ...... ---- 47 76 72 117 189
Av. final wt. of
S. nilotica
Males, g ........ ---- 365 466 435 387 294
Females, g ...... ---- 179 201 189 169 154
Av. both, g ...... ----.272 333 312 278 224
Production/ha of S.
nilotica (10O0 g +),
kg ............. 
4,944 2,351 
2,846 2,779 
2,480 1,998
Wt. of tilapia
marketable
(100g+)l,kg.... 0 95 94 96 99 83
Tilapia survival,
pct .................... 85 86 86 86 86
Predator survival,
pct .................... 57 62 75 64 56
Food conversion ... 2.4:1 5.1:1 4.1:1 4.2:1 4.7:1 5.8:1
Weight of tilapia marketable (percent)=
weight of tilapia over 100 g X 100
total weight 
of tilapia
The most serious current problem facing the development
of tilapia hybrid culture in the Northeast is the lack of proper
installations and techniques to produce adequate numbers of
hybrid fingerlings to meet the demand of commercial
growers. The problems in producing hundreds of thousands
of fingerlings were not fully appreciated until the demand by
private farmers increased rapidly.
Optimum numbers of male and female broodstock needed
to produce the greatest number of young have not been
determined. The greatest problem encountered is the large
variation in numbers of fry obtained between ponds stocked
with equal numbers of broodstock. Some ponds produce
several thousand fry while others produce only hundreds.
Occasionally, a spawning pond, drained after 3 months,
produces not one fry.
The exact reason for this large variation in fry production
is difficult to determine. The leading hypothesis is that the
male Sarotherodon hornorum and female Sarotherodon
nilotica are far enough apart taxonomically that their
spawning behavior differs to the extent that only a relatively
small portion of the sexually mature broodstock of the two
tilapia species successfully complete the spawning act.
Research is now underway to develop methods of producing
large numbers of tilapia hybrid fry with consistency.
Initial research showed that a ratio of 5 females to 1 male
tilapia gave results similar to that obtained when using 2
females to 1 male. Early research also demonstrated that
younger females and males, weighing less that 100 grams,
tended to produce more offspring than older, larger brooders,
and that 50 females and 10 males in a 350-square meter
spawning pond gave better results than lesser numbers of
brooders. Theoretically, young broodstock that have not
spawned before may form "families" that aid in the
spawning performance.
Experiments with Pirapitinga (Colossoma bidens)
4
Lovshin, Da Silva, Femrnandes, and Carneiro-Sobrinho (6)
found that pirapitinga has excellent culture potential
providing the fish could be induced to spawn under hatchery
conditions. Several earlier attempts to artificially spawn
pirapitinga proved unsuccessful. However, Da Silva and
Carneiro-Sobrinho (unpublished data) successfully spawned
this species in March 1976. Sexually mature fish of about 3
kilograms and 4 years of age were injected at 6-hour intervals
with pituitaries taken from curimata comun, Prochilodus
cearensis, Table 3. After the third injection, males and
females were placed together in a 4-square meter concrete
tank in the proportion of 2 males to 1 female. The fish
subsequently spawned in the tanks. The fertilized eggs were
collected and placed in incubators. Hatching success was
SOriginally described as Mylossoma 
bidens.
TABLE 3. SUMMARY OF THE DATA RELATING TO THE FIRST ARTIFICIAL
SPAWNING OF PIRAPITINGA (COLOSSOMA BIDENS) WITH INJECTIONS OF
CURIMATA COMUM (PROCHILODUS CEARENSIS) PITUITARIES
First Second Third
injection injection 
injection Spawned
Date................ 3/10/76 3/11/76 3/11/76 3/11/76
Hour ................ 9:00 p.m. 3:00 a.m. 9:00 a.m. 12:00 noon
No. of pituitaries in
2 cm' solution' ..... 12 12 18
Dosage per fish
Male (3.20 kg) ... 0.40 cm' 0.40 cm' 0.40 cm'
Male (3.40 kg) . . 0.75 cm' 0.75 cm 0.80 cm
Female (4.40 kg) .... 0.40 cm' 0.40 cm' 0.40 cm'
Donor fish averaged 500 g weight.
high but all the fry died within 
3 days. Exact causes of this
mortality are not known. 
A second attempt to spawn
pirapitinga 2 weeks later using 
the same technique resulted 
in
a partial spawn, Table 4. 
Fry from the latter spawn 
were
taken to a DNOCS fish hatchery 
with better nursery
conditions. Of approximately 
3,000 fry, 156 survived 
to
fingerling size. This is the 
first time that pirapitinga has 
been
successfully spawned and 
raised to fingerling size. 
With
improved spawning and 
rearing techniques, large 
numbers
of pirapitinga fingerlings 
can probably be produced 
on a
regular basis.
TABLE 4. SMMARY OF TIlHE DATA 
RELATING TO THE SECOND ARTIFICIAL
SPAWNING OF PIR A'ITINCGA (COLOSSO.IA 
BIDENS) 'AWITH INJECTIONS OF
CO RIMASTA COMUM (PRoCHILODUS (CEARENSIS) 
PITUII ARIES
First Second
injection injection
Third
iniection
Spawned
Date ................. 4/6/76 
4/6/76 4/7/76 4/7/76
Hour ................ 6:00p.m. 
12midnight 6:00a.m. l1:15a.m.-
1:00 p.m.
No. of pituitaries 
in
I cm
r
' solution . . ... 12 
12 24
Dosage per fish
Male (3.20 kg) ...... 0.30 m 
0.30 c
r
m 0.30 cmi
Male (2.20)......... 0.30 cm' 
0.30 em' 0.30 cm
Female (2.68 kg) .... 0.40 cm' 0.40 cm' 0.10 em:
' Donor fish averaged 500 g weight.
Experiments with Tambaqui (Colossoma 
macropomum)
Research by Lovshin, Da Silva, Fernandes, 
and Carneiro-
Sobrinho (6) demonstrated that tambaqui also 
has excellent
culture potential if this species can be 
spawned to provide
fingerlings for stocking purposes. To 
date, a number of
spawning attempts over a 3-year period have 
been unsuc-
cessful although several abortive spawns have 
been obtained
using pituitary injections from the donor 
fish, Prochilodus
cearensis.
Experiments with Curimata Comum
(Prochilodus cearensis)
Curimata comum, a fish native to Northeastern 
Brazil, for
mans years has been artificially spawned and 
raised to fin-
gerling size in DNOCS fish hatcheries 
and subsequently
stocked into reserxoirs. This fish is readil% accepted 
b loc al
consumers, making it a good candidate species 
for intensive
fish culture.
Research was carried out by DNOCS 
biologists to deter-
mine the growth and production of curimata 
comum at a
Originally described as Colossoma bidens.
Pirapitinga (Colossoma bidens) is one of several Brazilian species
being tested for culture potential at Pentecoste. This fish has 
been ar-
tificially spawned and promises 
to be an excellent alternative 
to the
tilapia hybrid culture system.
number of different stocking rates. 
Ten earthen ponds of 355
square meters were stocked with curimata comum at levels
ranging from 1,000 to 10,000 
per hectare. Fish were fed
agricultural waste products, 
principally wheat bran and
babacu cake, at 5 percent 
of their body weight daily, 6 
d(lays a
week. Fish were sampled monthly 
and new feeding rates
were calculated from sample 
results. After 187 days, ponds
were drained and all fish were 
counted and weighed. Produc-
tion was poor at all stocking 
rates. Curimata stocked at the
lowest densities did not attain 
a marketable average weight
of 100 grams after 6 months, 
and the maximum production
per hectare was 486 kilograms. 
Conversion rates were poor,
ranging from a low of 7.0 to I to 
a high of 11.3 to 1. Data in
Table 5 indicate that curimata comum 
is not an acceptable
fish for intensive culture using 
the culture technique applied
in this experiment.
TABLE 5. ScUsIAM A OF RESL ITS OF PHOD (C 
IION OF C HAIS) Ai COM I :M (PROC(III()UoDS CEARENSIS)
IN EsRTHE PONDS ST VARIOL S LE\ l s OF SI ocKIN;
Performance measure
F ish/pond ........ ........
As. initial weight, g .........
Av. final weight, g .........
Production/pond, g ..........
Production/ha, kg ...........
Su r i al, p et ................
Quantity of ration fed, g .....
Feed cons ersion ............
No. of days in experiment....
Result, bs stocking rate per hectare
1,000 2,000
35 71
11 18
95 60
2,649 4,150
74 116
80 97
19,095 24,920
8.4 8.7
187 187
' Does not include 12 kg of tilapia found in pond at harvest.
3,000
106
11
74
7,430
208
94
43,630
7.0
187
4,000
142
14
85
11,190
313
92
67,655
7.3
187
5,000
177
12
63
10,330
289
93
68,240
8.3
187
6,000
213
13
69
14,170)
397
96
91,770
8.1
187
7,000
248
13
71
17,344
486
98
98,460
7.0
187
8,000
284
13
53
14,504
406
96
90,180
8.3
187
9,000
319
12
45
14,130
396
98
96,510
9.4
187
10,000
355
13
37
12,026
337
93
84,090
11.3
187
L
------
L
Experiments with Curimata Pacu (Prochilodus argenteus)
Native to the Sao Francisco River, curimata pacu was
introduced into Ceara by DNOCS biologists and is now being
artificially spawned and raised to fingerling size for
distribution in local waters. While similar to curimata
comum in appearance and life history, curimata pacu grows
to a much larger size. For this reason, curimata pacu was
tested by DNOCS fishery biologists for growth and produc-
tivity in six 355-square meter earthen ponds at various levels
of stocking. Fingerlings were stocked at 1,000, 3,000, 4,000,
5,000, 7,000, and 9,000 per hectare and fed babacu cake at 5
percent of their weight daily, 6 days a week. Feeding rates
were readjusted using monthly seine samples. After 301 days,
the ponds were drained and all fish were counted and
weighed. Fish at all stocking rates reached a marketable size
of 100 grams, Table 6; however, food conversion rates
ranged from 5.8 to 1 to 8.7 to 1. A comparison of the average
weights of curimata comum at harvest and curimata pacu
after 6 months demonstrates that curimata pacu grows more
rapidly under similar cultural conditions and average initial
weights. Based on this experiment, curimata pacu still cannot
be considered a good fish for intensive culture.
TABLE 6. SUMMARY OF RESULTS OF PRODUCTION OF CURIMATA PACU
(PROCHILODUS AR GENTEUS) IN EARTHEN PONDS AT VARIOUS LEVELS
OF STOCKING
Performance Result, by stocking rate per hectare
measure 1,000 3,000 4,000 5,000 7,000 9,000
Fish/pond ........ 35 105 140 175 245 315
Av. initial wt., g ... 14 12 13 14 12 13
Av. wt. at 6 mo., g. 149 96 156 96 96 83
Av. final wt., g .... 199 104 158 153 120 145
Production/pond, g 6,366 10,468 26,134 25,595 23,041 34,227
Production/ha, kg 178 293 732 717 645 958
Survival, pet ...... 91 95 100 95 78 75
Quantity of ration
fed, g .......... 45,650 88,750 188,290 149,410200,430248,180
Feed conversion... 7.2 to1 8.5 to1 7.2 to1 5.8 to1 8.7 to1 7.3 tol1
No. of days in
experiment ...... 301 301 301 301 301 301
Experiments with Channel Catfish (Ictalurus 
punctatus)
In October 1972, channel catfish fry were introduced to
Northeast Brazil from the United States after having been
examined and found free of parasites and disease. At present
120 broodstock 4 years old and averaging about 4 kilograms
each are divided equally between two 5,000-square meter
earthen ponds. The ponds contain tilapia and shrimp as
forage for the broodstock and the fish receive a pelleted
ration daily. Growth of the fry received from the United
States has been excellent due to favorable climatic and water
quality conditions. The ponds are equipped with spawning
boxes. Two spawns to date have been obtained from one
pond and none from the second pond. This poor spawning
record substantiates observations from other tropical regions
that channel catfish are unable to consistently spawn
naturally in warm, tropical waters, which probably makes
this species an unlikely candidate for intensive culture in
Northeast Brazil.
FISH CULTURE EXTENSION
There are now 16 farmers raising fish in the Northeast,
most located in the state of Ceara. The total pond area in
water is about 5 hectares. There is strong farmer interest in
raising tilapia, and the demand for hybrid fingerlings is
TABLE 7. COST-BENEFIT ANALYSIS OF THE INTENSIVE CULTURE OF TILAPIA
HYBRIDS IN A 0.8-HECTARE POND ON THE PASSAGEM FRANCA FARM
Item Amount
1. Costs
Fixed costs
Guard service .................. Cr$ 600.00
Pond maintenance .............. 200.00
Total fixed costs ................ Cr$ 800.00
Variable costs
Wheat bran (1,120 kg) ........... . 550.40
Chicken manure (6m
3 ) 
........... 450.00
Harvest labor .................. 400.00
Total variable costs .............. Cr$ 1.400.40
TOTAL COST ................... .. Cr$ 2,200.40
2. Revenues
Gross income (1,988.50 kg at Cr$5.50) Cr$10,936.75
Net income ....................... .Cr$ 8,736.35
($1,027.69 U.S.)
3. Construction of ponds' ............. Cr$ 5,590.35
' Cost of pond construction could be paid for with profits earned in
culture period.
greater than the supply. This is the principal problem facing
the expansion of intensive fish culture. Farmers that are
raising fish are producing large crops of hybrids resulting in
high economic returns. Carneiro-Sobrinho, Fernandes, and
Jensen (unpublished data) made an economic study of two of
the largest farms in the region. One of these, Passagem
Franca, has been farming fish for over 2 years. The first crop
of tilapia hybrids produced in the farm's 8,000-square meter
semi-natural pond fully paid pond construction costs and still
left the farmer with a profit of Cr$3,146, equivalent to $370
U. S. (4). This pond was constructed by damming a valley
below a farm reservoir and using water from the reservoir to
fill the fish pond. The pond was stocked for the second crop
with 8,000 tilapia hybrids having an average weight of 13
grams. The fish were fed daily a ration of wheat bran at 3
percent of total weight of fish in the pond. This ration was
supplemented with liberal additions of cow and chicken
manure.
After 5 months, fish were harvested daily over a 2-month
period. Harvest was by cast-net, nook and line, and draining.
Average weight of the hybrids was 283 grams and the total
production was 1,989 kilograms (2,486 kilograms per
hectare). Survival of tilapias was calculated at 87 percent.
Profit for the 7-month period was Cr$8,735.00 ($1,027.69
U.S.), Table 7.
TABLE 8. COST-BENEFIT ANALYSIS OF THE INTENSIVE CULTURE OF TILAPIA
HYBRIDS IN TWO PONDS TOTALING 1.5 HECTARES ON THE TRAPIA FARM
Item Amount
1. Costs
Fixed costs
Guard service ................... Cr$ 1,000.00
Harvest equipment (nets) ......... .. 200.00
Pond maintenance costs ............ 200.00
Total fixed costs 200..Cr$ 1,400.00
Variable costs
Harvest labor .................. 700.00
Total variable costs ............. Cr$ 700.00
TOTAL COST ........... Cr$ 2,100.00
2. Revenues
Gross income (3,520 kg at Cr$ 5.50).. Cr$19,360.00
Net income .. ....................... Cr$17,260.00
($2,031.00 U.S.)
3. Construction of ponds' ...... Cr.$ 6,000.00
I Cost of pond construction could be paid for with profits earned
in first culture period.
9
A second farm, Trapia, included two semi-natural ponds of
5,000 square meters and 10,000 square meters. These ponds
were also constructed by building a series of earthen dams
below a farm reservoir. The smaller pond was stocked with
5,000 tilapia hybrids averaging 25 grams and the larger one
received 5,000 fingerlings of 26-gram average weight. The
main farm enterprises include milk cows, chickens, and pigs.
Waste products from these animals were used to fertilize the
ponds. Each pond received an initial application of 500
kilograms of chicken manure before stocking the fish and
then 500 kilograms of chicken manure every 15 days until
harvest. A pig sty was placed on the margin of each pond; the
smaller pond had a sty with 15 pigs and the larger one had a
sty with 16 pigs. The pigs were fed the waste curds from a
cheese making process that utilized the milk from the cows.
The pig waste products were washed into the ponds daily. No
other ration was used and the manures added periodically
cost nothing.
After 6 months of growing time, the ponds were harvested
daily over a 2-month period by cast-net, hook and line, and
draining. The fish from both ponds averaged 400 grams and
the total production was 3,520 kilograms (2,347 kilograms
per hectare) from both ponds. Even when pond construction
costs were deducted, the net profit from these two ponds was
Cr$11,260 ($1,324.70 U. S.), Table 8.
Two fish culture projects have been initiated in DNOCS
irrigation projects. A pond of 1.2 hectares has been built and
stocked in the state of Paraiba and a second pond of 0.7
hectare has been stocked in the state of Pernambuco. Before
the end of the year, seven ponds consisting of 2 hectares of
water should be completed and in production at irrigation
projects in the state of Ceara.
10
BIBLIOGRAPHY
(1) DAVIEs, W. D. 1972. Progress Report on Fisheries 
De-
velopment in Northeast Brazil, II. Project AID-2270,
Task Order 4, International Center for Aquaculture,
Auburn Univ., Auburn, Ala.
(2) JEFFREY, N. B. 1972. Progress Report on Fisheries De-
velopment in Northeast Brazil, I. Project AID/csd-2270,
Task Order 3, International Center for Aquaculture,
Auburn Univ., Auburn, Ala.
(3) JENSEN, J. W. 1974. Progress Report on Fisheries De-
velopment in Brazil. Project AID-2270, Task Order 8,
International Center for Aquaculture, Auburn Univ.,
Auburn, Ala.
(4) . 1975. Report on the Intensive Culture 
of
the Tilapia Hybrid (Tilapia nilotica X Tilapia horn-
orum) on a Private Farm in Northeast Brazil. Centro
Technique Forestier Tropical-Nouvell 
Serie, No. 10,
p. 27-29.
(5) 1976. Progress Report on Fisheries De-
velopment in Northeast Brazil. Project AID-1152, Task
Order 2, International Center for Aquaculture, Auburn
Univ., Auburn, Ala.
(6) LovsmHIN,, L. L., A. B. DASILvA. J. A. FERNANDES, AND
A. CARNEIRO-SOBRINHO. 1974. Preliminary Pond Culture
Tests of Pirapitinga (Mylossoma bidens) and Tambaqui
(Colossoma bidens) from the Amazon River Basin. FAO
Aquaculture Conference for Latin-America. Montevideo,
Uruguay.
(7) , A. B. DASILVA, AND J. A. FERNANDES. 
1974.
The Intensive Culture of the All Male Hybrid of 
Tilapia
hornorum (male) X Tilapia nilotica (female) 
in North-
east Brazil. FAO Aquaculture Conference 
for Latin-
America. Montevideo, Uruguay.
(8) . 1975. Progress Report on Fisheries De-
velopment in Northeast Brazil. Project AID/csd-2270,
Task Order 8, International Center for Aquaculture,
Auburn Univ., Auburn, Ala.
(9) AND A. B. DASILVA, 1975. Culture of Mon-
osex and Hybrid Tilapias. FAO/CIFA 
Symposium on
Aquaculture in Africa, Accra, Ghana.
For more information communicate directly with the
Director of the Department of Fisheries and 
Fish Culture at
the following address:
Director
Departmento Nacional de Obras
Contra as Secas (DNOCS)
Av. Duque de Caxias, 1700
Caixa Postal 650
60.000 -Fortaleza, Ceara, Brazil
11