Progress Report on Fisheries Development
in Northeast Brazil
k
INTERNATIONAL CENTER FOR AQUACULTURE
AGRICULTURAL EXPERIMENT STATION 
AUBURN UNIVERSITY
R. DENNIS ROUSE, DIRECTOR AUBURN, ALABAMA
RESEARCH AND DEVELOPMENT SERIES NO. 26 Project: 
AID 1152T.O.2. FEBRUARY 1980
CONTENTS
Page
TRAINING AND TECHNICAL ASSISTANCE ............................ 
3
INSTALLATIONS AND FACILITIES .................................. 4
FISH CULTURE RESEARCH PROGRAM .............................. 4
Tilapia Production Experiments ................................ 4
Production of All-male Tilapia Hybrid Fingerlings.............7
Evaluation of Tilapias in Intensive Fish Culture................9
Experiments with Tambaqui (Colossoma macropomum) and
Pirapitinga (Colossoma bidens) ............................. 9
RESERVOIR FISHERIES .......................................... 11
FISH CULTURE EXTENSION ..................................... 12
BIBLIOGRAPHY ................................................ 15
PUBLISHED FEBRUARY 1980-2M
COVER PHOTO. Harvesting all-male tilapia hybrid fingerlings at the
Valdemar C. de Franca fish hatchery in Maranguape, Brazil, for
stocking intensive fish culture ponds.
Information contained herein is available to all without
regard to race, color, or national origin.
Progress Report on Fisheries Development
in Northeast Brazi
LEONARD L. LOVSHIN
1
AUBURN UNIVERSITY, under contract by USAID (U.S.
Agency for International Development), has been involved in
improving the freshwater fishery resources in Northeast Brazil
since 1966 when several staff members participated in short-
term work programs in that country. From 1969 to the present,
the University has provided technical assistance in the form of
resident advisors to the Brazilian federal agency, DNOCS
2
, in
fish culture, reservoir management, and fish culture extension.
The author was the resident fish culture advisor from June 1972
until his return to Auburn in January 1979. The fisheries technical
assistance contract to the Brazilian government was terminated
on February 28, 1979. Principal duties involved assisting Brazil-
ian biologists in conducting and analyzing fish culture research
and aiding DNOCS in planning and implementing semi-annual
fishery short courses for Brazilian and other Latin American
fishery workers.
This report summarizes the pertinent activities related to
fishery and fish culture research and training within the DNOCS
fishery organization during the period January 1, 1977 - Decem-
ber 31, 1978. Previous progress reports describe fisheries
development in Northeast Brazil (3, 7, 8, 9, 10, 11).
TRAINING AND TECHNICAL ASSISTANCE
The Centro de Pesquisas Ictiologicas is recognized as one of
the outstanding tropical freshwater fishery research and
training organizations in Latin America. Staff of the research
center are frequently invited to present research and position
papers at national and international fishery conferences. The
number of requests for technical assistance in fisheries and
aquatic biology from Brazilian and other South American
agencies is growing yearly. DNOCS biologists have assisted
Brazilian state and federal agencies and academic institutions in
planning and carrying out a wide range of fishery activities to
develop Brazil's fishery resources and trained manpower.
One of the biologists from the Centro de Pesquisas
Ietiologicas teaches advanced aquaculture and directs under-
graduate thesis candidates at the local Federal University of
Ceara. He also acts as a coordinator for the fish culture section of
"Projecto Ser/anejo," and a cooperative rice-fish culture re-
search and development project with CODEV\ASF
3
. The
"Projecto Sartanejo" is a large federal project that finances
construction of small earthen reservoirs throughout Northeast
Brazil. The impoundments will be used to establish integrated
farming activities, including fish culture. CODEVASF is one of
the Brazilian governmental agencies responsible for the
development of the Sao Francisco River valley. CODEV\ASF
has large irrigated areas in rice cultivation and is interested in
Associate Professor, Department of Fisheries and Allied Aqua-
cultures.
2
Departamento Nacional de Obras Contra as Secas.
:Companhia de Desenrolvimento do Vale do Sao Francisco.
integrating operation of these areas with fish culture. Brazilian
extension agents and researchers connected with both DNOCS
and CODEVASF are being trained at DNOCS fishery installa-
tions.
A formal agreement between Brazilian and Peruvian govern-
ments to exchange technical expertise has involved the research
center. A DNOCS fish culture specialist traveled to Peru to help
Peruvians establish and improve their own freshwater fishery
programs based on the Brazilian experience in the Northeast.
Peruvian fishery scientists receive training at the Centro de Pes-
quisas Ictiologicas in areas of aquatic resource management that
will enhance the Peruvian program. In 1978, three biologists
from the Peruvian Ministry of Fisheries spent 9 weeks at the re-
search center.
The Centro de Pesquisas Ictiologicas continues to offer short-
term training programs in the aquatic sciences to interested
Latin American biologists. Increased demand for training has
resulted in the Centro de Pesquisas Ietiologieas offering two for-
mal training courses a year. Training programs lasting 2 months
are offered, March-April and September-October, for up to 15
students. These short courses contain classroom lectures and
practical exercises in the subjects of intensive and extensive fish
culture, reservoir management, collection of commercial fish-
ery statistics, hatchery management, and reservoir limnology.
Short-term training was offered to 6 Hondurans, 7 Colombians,
1 Panamanian, 1 Venezuelan, and 41 Brazilians in 1977 ad -978.
These Honduran biologists participated in a 2-month fishery train-
ing program at the Rodolpho von Ihering fish culture center in
Pentecoste, Brazil. The training program is offered twice a year 2o
Latin American biologists.
Officials of the Brazilian government 
are aware of the
contribution that the DNOCS fishery programs are has ing 
on
animal protein production in the northeast 
of Brazil. DNOCS is
considered the leader in freshwater fishery 
research and
commercial fish production wsithin Brazil. 
DNOCS's elforts to
train Brazilian and other Latin American biologists 
at the Centro
de Pesquisas Ictiologicas have also 
been well received. Realizing
that the Ccntro de Pesquisas Ictiologicas 
requires strengthening
to continue its leadership 
role in Brazil, the federal 
government,
through the Conselho Nacional des Pesauisas, 
has granted the
research center Cr. $9 million (U.S. 'W00,000) for 1979. 
These
additional funds wvill allow the research center to hire 
and train
new biologists to aid the present stafi iii meeting 
the numerous
requests for technical assistance, 'training, 
and expanded
research activities.
INSTALLAT0ONS AND FACIL IES
News headquarters of the Centro dc Pesquisas 
Ictiologicas iS
scheduled to he inaugurated in 1979. The center is shifting 
to
Pentecoste, Ceara, 100 kilometers by paved 
road from the state
capital of Fortaleza. The new installations 
contain a 1,200-
square meter laboratory and teaching complex, administration
building, warehouse, 
garage-machine shop, 
fish technology
building, and net fabrication and repair house. 
The building
complex is surrounded by pond facilities on 120hectares 
of land
located belows the 5,000-hectare Pereira de Miranda 
Reservoirat
Pentecoste. Housing facilities for station biologists, 
visiting
scientists, and students are 
near completion. Presently, the
Centro de uPesquisas Ictiologicas has 162 earthen 
ponds and 36
concrete ianlks ranging in size ftom 100 
to 10,0Cf' square meters
and 36 to 75 square meters, respectively. 
A wx et-laboratory
containing forty 20-iter aquaria and txxents 500-liter 
cem nt
fiber tanks plus eight concrete tanks for holding 
fis is available.
Total water area is 11.0 hectares.
When ,)ond installations are completed 
in 1980, the Centro de
Pusquisas ctiologicas will contain over 200 carthen, 
ponds
comprising more than 20 hectares of water. The research
center's Facilities are among the finest 
in the tro ical sw-orld for
research and training in freshwater fish 
culture, reservoi
management, reservoir linnology, and related 
aquatic sciences.
FISH CLTURE 'RESEARICH PO4A
A(cus cltural research has seen in progress 
since 1970,
directed Itoarcis evxabating Brazilian fishes for culture potenial
and developing culture systems for use in DNOCS 
irrigation
projects and by private farm es. Exotic species with known
extensiv'e and intensive culture potential are being investigated.
This section summarizes pertinent research performed 
over the
past 2 xears. Evaluations of the most promising culture soecies
'ased on 7 -,ars of research are included.
Thpi Podtclbon E'zpertme'ls
At present, tilapia offer the greatest totential for iemediate
culture in ponds in Northeast Brazil. Past re se'arch r sults can be
Phase I ol Rodolpho von ihering tish cuure research center in Pen'ecose, Brazil, provides 
pond iacilities for ,ish culnure 2raining anr -e-
search. Note pig sties constructed over the ponds in loxer lt corner ol photograph.
Phase O of the Rodolpho von lhering fish culture center is scheduled 
for completion in 1980. The complex will include teaching, research,
administrative, storage, repair, and dormitory facilities. An aerial view of the center is at top, with building detail shown bottom left and wet
laboratory research installations shown bottom right.
reviewed in publications by Lovshin, Da Silva, and Fernandes
(13), Lovshin (10, 11), and Lovshin and Da Silva (14). The
majority of research on tilapias is presently concentrated on the
all-male tilapia hybrid produced by crossing female Tilapia
nilotica with male Tilapia hornorum.
Tilapia hybrids give excellent results when intensively
cultured utilizing low-cost agricultural by-products or organic
manures as feeds and fertilizers. Past research demonstrated that
tilapia hybrids stocked at 10,000 per hectare and fed agricultural
by-products reached an average weight of 200-300 grams in 6
months, resulting in 2,000 to 3,000 kilograms per hectare.
Maximum production of over 10,000 kilograms per hectare per
year of 400-gram fish was obtained by stocking up to 31,000
male tilapia hybrids per hectare and feeding cottonseed and
palm nut cake.
Piava, Carneiro-Sobrinho, and F. Melo (unpublished data)
tested the growth of the all-male tilapia hybrid fed three types of
locally available agricultural by-products. A completely random
design was used with three treatments replicated three times.
Nine 355-square meter earthen ponds were stocked with 400
tilapia hybrids (11,250 per hectare). Castor bean meal (30
percent protein), babacu cake (a palm nut with 21 percent
protein), and cottonseed cake (21.5 percent protein) were fed at
3 percent of the averagebiomass of fish in each treatment once a
day, 6 days a week. Every month at least 10 percent of all fish
were seined from each pond, weighed, and measured. Feeding
rates were recalculated monthly based on the seine samples.
After 238 days all ponds were harvested. Results of this
experiment can be found in table 1. Differences in total fish
production between treatments were significant at the 0.05
TABLE 1. S M MAI OF s{ 1 IS COM PAINi TINR' TI PEiS OF FtIEDS ON,
I't, Gionii o Ai -MAE TILAPIA fIsYBIDS (FENIiALp
T. N5LOTICA x MAL .IiOrNOI M)
Performance measure
Av. initial w t., g ..........
Av. final w t., g ...........
Av. production per
pond, kg ...............
Av, production 
per
ha, kg . ..... .... ........
Survival pet .............
Av. conversion rate
6 m onths...............
7 m onths...............
8 m onths ...............
Growth of fish per
day, g .................
Result, by type of feed'
Castor bean Palm nut Cottonseed
meal cake cake
15 14 15
323 266 239
123 93 88
3,444
95
1.8:1
2.0:1
2.4:1
2,325 2,200
95 93
1.3 1.1 0.9
lAverage of three replicates.
level. Fish fed castor bear mea
_ 
gained siiany more
weight than fish fed babacu or cottonseed cake. No significant
difference in fish production was found between fish fed
babacu cake or cottonseed cake. Castor bean meal (80.10 per
kilogram) proved to be the most profitable feed although
babacu cake (80.11 per kilogram) and cottonseed cake ($0.15
per kilogram) produced fish economically. Feed conversions
(kilograms of feed per kilogram of fish growth) for castor bean
meal, babacu cake, and cottonseed cake were 2.4, 2.6, and 2.7,
respectively.
Augusto and H. Melo (unpublished data) tested effects of
ammonium sulfate (20 percent N), triple superphosphate (46
percent P
2
0s), ammonium sulfate plus triple superphosphate,
and no fertilizer (control) on the growth of the all-male tilapia
hybrid in 355-square meter earthen ponds. Treatments receiving
ammonium sulfate and triple superphosphate were replicated
three times while treatments receiving both fertilizers and the
TOP: Monthly seine samples are taken to follow the growth of ex-
perimental pond cultures at the Rodolpho von Ihering fish culture
center. LEFT: Fish taken from the experimental ponds are weighed
and measured by DNOCS biologists to evaluate production
practices.
control were replicated twice. Ponds in the ammonium sulfate
and triple superphosphate treatments were fertilized with equal
amounts of nitrogen and phosphate (280 grams per application).
Ponds in the ammonium sulfate plus triple superphosphate
treatment were fertilized with half the amount of nitrogen (140
grams per application) and phosphate (140 grams per
application) so that the total amount of nitrogen plus phosphate
applied was equivalant to 280 grams. Ponds were fertilized
every 15 days by placing the fertilizer on a platform located 20
centimeters below the water surface. Ponds fertilized with
ammonium sulfate and triple superphosphate received 1.42
kilograms per pond per application (40 kilograms per hectare)
and 0.61 kilogram per pond per application (17 kilograms per
hectare), respectively. Ponds receiving both fertilizers had 0.76
kilogram per pond application of triple superphosphate (20
kilograms per hectare) and 0.31 kilogram per pond application
of triple superphosphate (8.7 kilograms per hectare) added. All
fertilized ponds received an initial application of fertilizer 15
days before fish were stocked.
Ponds were stocked with the equivalent of 10,000 all-male
tilapia hybrids per hectare. All ponds were sampled once a
month. Ten percent of the fish in each pond were captured with
a seine, weighed, and measured. Three fish from each pond
were sacrificed monthly and stomach contents analyzed. After
192 days, the experiment was terminated and all ponds were
harvested.
Results of this research are summarized in table 2. The use of
chemical fertilizers increased the total production of tilapia
hybrids two to three times that of the control. Triple superphos-
TABLE 2. SUMMARY OF RESULTS OF THREE TYPES OF CHEMICAL FERTILIZERS AND A CONTROL ON THE PRODUCTION OF
ALL-MALE TILAPIA HYBRIDS (FEMALE TILAPIA NILOTICA x MALE TILAPIA HORNORUM),
Result, by fertilizer treatment'
Performance measure Control Ammonium Triple super- Ammonium sulfate plus
sulfate phosphate triple superphosphate
Av. initial wt., g .............................. 20 19 19 20
Av. final wt., g ............................... 43 . 118 101 122
Av. production per pond,
2 
kg .................. 12.5 36.3 29.7 34.9
Av. production per hectare, kg................. 350 1,016 832 977
Survival, pct.
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
81 87 82 81
Fertilizer per pond, kg ........................ 0 22.7 9.8 16.3
Fertilizer per hectare, kg ...................... 0 636 274 456
'The control and ammonium sulfate + triple superphosphate treatment results are the average of two replicates. The remaining two treatments are the
results of three replicates.
2
Weight of fish at final harvest; sacrificed fish not included.
3
Survival of fish at final harvest; sacrificed fish not included.
phate alone was not as effective as ammonium sulfate alone or
the two fertilizers together. Apparently, a source of nitrogen is
needed to increase fish production even in older earthen ponds.
Stomach analyses demonstrated that the all-male tilapia
hybrid fed almost exclusively on planktonic and attached algae
and bottom muds. Although ponds were rich with zooplankton,
benthic organisms, and aquatic insects, few of these items were
encountered in stomachs of the hybrids.
Production of All-male Tilapia Hybrid Fingerlings
Several researchers have reported producing all-male
offspring when female Tilapia nilotica are crossed with male
Tilapia hornorum. Lovshin, Da Silva, Carneiro-Sobrinho, and F.
Melo (unpublished data) performed a series of experiments to
determine a system for producing maximum numbers of all-
male hybrid fingerlings. Tests were performed to determine the
optimum ratio of female T. nilotica to male T. hornorum,
number of female T. nilotica per area of spawning pond, and
length of time broodstock can be used before replacement is
necessary. A total of 142 crosses was made over a 6-year period.
Original stocks of T. hornorum and T. nilotica were obtained
from Ivory Coast, West Africa, and the offspring of these stocks
were used as broodstock for the experiments. Mature female T.
nilotica and male T. hornorum were stocked into the hybrid
spawning ponds in varying numbers per surface area and at
different ratios. The broodstock were fed agricultural by-
products at 5 to 10 percent of their body weight, 6 days a week.
After 70 to 92 days, the ponds were drained and all fingerlings
recovered were counted. The broodstock were counted and
weighed individually and immediately transferred to a newly
prepared spawning pond where a new hybridization study was
initiated.
Table 3 has the results of fingerling production when 25, 50,
75, and 100 female T. nilotica were stocked in ponds of 350
square meters. The ratio of female T. nilotica stocked with the
male T. hornorum was 5 to 1 in most trials. Best results were
obtained with 50 females, or 1 female per 7 square meters of
TABLE 3. SUMMARY OF THE RESULTS OF INCREASING THE NUMBER PER
AREA OF FEMALE TILAPIA NILOTICA AND MALE TILAPIA HORNORUM
ON THE PRODUCTION OF ALL-MALE TILAPIA HYBRID FINGERLINGS
Result, by male:female numbers
Performance measure 5:25 
10:50 15:75 20:100
No. of replications ..... 15 61 17 9
Av. No. of fingerlings... 1,179 2,763 2,167 1,502
Range of av. No. ofi
fingerlings ........... 0-3,309 0-8,443 473-5,295 607-2,526
Av. No. of fingerlings/
female .............. 52 57 30 15
spawning pond. An average of 2,763 hybrid fingerlings was
produced per 2.5-month spawning period per pond.
Data in table 4 show how varying the ratio of female T. nilo-
tica to male T. hornorum affected tilapia hybrid fingerling
production in 350-square meter earthen ponds. The commonly
used female to mal
- 
ratio of 5 to 1 was tested against ratios of 1
female to 1 male and 1 female to 2 males. Increasing the number
of male T. hornorum to female T. nilotica increased the number
of hybrid fingerlings produced. Best results were obtained when
two T. hornorum males were stocked for every T. nilotica
female.
TABLE 4. SUMMARY OF THE RESULTS OF INCREASING THE RATIO OF FEMALE
TILAPIA NILOTICA TO MALE TILAPIA HORNORUM ON ALL-MALE
TILAPIA HYBRID FINGERLING PRODUCTION
Performance measure
Av. initial wt., g (males) ............
Av. initial wt., g (females) ..........
Av. No. of 
fingerlings 
.............
Av. No. of fingerlings/female
2 
......
Av. No. of days in period ..........
Result, by male:female ratio'
5:25 25:25 50:25
103 144 101
103 105 101
699 1,532 2,159
30 64 88
80 82 80
'Average of six replicates.
2
In some cases, less than 25 females were found at draining.
Table 5 gives results of the effects of repeated use of the same
group of broodstock (10 T. hornorum males and 50 T. nilotica
females) on fingerling production. Groups of broodstock were
repeatedly transferred to freshly prepared 350-square meter
spawning ponds until fingerling production declined, at which
point the broodstock were removed. This procedure gave an
indication of the length of time the brooders could be used
before replacement with new broodstock. After the third
spawning period, the production of hybrid fingerlings declined.
Therefore, after approximately 8 months or 3 spawning periods,
TABLE 5. SUMMARY OF RESULTS OF REPEATED USE OF BROODSTOCK
GRouPS, 50 FEMALE TILAPIA NILOTICA x 10 MALE TILAPIA
IIORNORUM, ON ALL-MALE TILAPIA HYBRID FINGERLING
PRODUCTION
Pformance measure 
Result, by spawning 
period
Performance measure 1 2 34
1 2 3 4
No. of replicates ....... 12 12 12 6
Av. No. of fingerlings . 2,969 2,405 2,569 1,465
Range of av. No. of
fingerlings ........... 620-6,000 412-4,679 0-8,443 0-2,976
Av. No. of fingerlings
/female .............. 57 50 53 33
Range of av. No. of
fingerlings/female .... 12-120 8-61 0-192 0-69
TOP: The Valdemar C. de Franca fish hatchery, one of live hatch-
eries operated by DNSOCS in the Northeast of Brazil, produces
fingerlings for stocing into public and private reservoirs. RIGHT:
Fingerlings being harvested.
the broodstock would be removed and new broodstock
introduced. At the time of replacemenut, brood stock are about 13
to 14 months old assuming th~ey -were stocked when sexually
mature at 5 to 6 months (60 to 100 grams).
The drawing illustrates the procedure now used by DN GUS
to produce all-male tilapia hybrid fingerlings. Pure stocks of T.
hornorum and T. nilotica are held in covered 36-square meter
concrete tanks w ith a filtered wxater supply. Pure hroodstock of
both sp~ecies are released into 330-square meter ponds separated
by at least 50 meters to prevent contamnination. These earthen
ponds are used to produce the number of T. nilotica and T.
hornorum broodstock required for mass producing all-male
hx brids. XX hen the imimature, pure stock fingerlings reach a size
at which they can be readily sexed (20 to 30 grams), female T.
nilotica and male T. hornorum are placed in separate ponds to
allow them to grow to maturity isolated from their opposite sex.
If pond facilities are limited, fish can he stocked into cages and
fed a complete ration. Immature female T. oilotica and male T.
hornorum should re~ach sexual rnaturity in 2 months, 60 to f00
gramns, when stocked two to three per square meter and fed 5
percent of their body weight daily. Mature male T. horoorum
and female T. nilotica with sw ollen genital papillae are p~laced in
the hybridization pond at the ratio of one male to one female.
stocking one female per 7square meters ot pond surface area.
WXater depth of spaw ning ponds should be slightly less than 1
meter. After 2.5 months, by brid fingerlings should be removed
from the spawxning pond to axvoid possible backerossing wxith the
parents. Hybrid fingerlings should be separated from by brid fry
to reduce cannibalism.
Fingerling production remains the primary constraint to
raising all-male tilapia hy brids on a commercial scale. Fingerling
production pe~r scuare mecter of spawxning pond is low . Reasons
for low hy hbrid fingerling prod iictix itx are not thoroughly
understood. Apparently, the percentage of female T. oilotica
that xx ill spawn wxith mnale T. hornoum is loxx. This appears to he
caused by differences in spa-x'ning behavior betxx eon the txx ,o
sp~ecies. Femnales that do spaxwn produce normal numbers of
by brid fry . At oresent, a large pond hatchery facility wxould he
required to produce the number of all male hy bridi fingerlings
necessary for a large groxx out operation.
Evaluaflon of Tilaplas in lntensve Fls Culture
Use of tdlapias in intensiv e fish culture has been highly
promising wxhen some method of reproduction control is
System for producing all-male Tilapia hybrid 
fingerlings (female
Tilapia niiotica x male Tilapia hornorum).
practiced. Reproduction control 
with a predator (11) and
culture of selected males (13, 14) or all-male 
tilapia hybrids (10,
11, 14) have all successfully reduced 
or eliminated excessive
reproduction in Brazil. Research has been 
concentrated on the
culture of all-male hybrids. The 
all-male tilapia hybrid has
proven to be an exceptional culture 
fish, able to withstand
handling, diseases, and low levels of 
dissolved oxygen in culture
ponds. Farmers in demonstration trials 
have harvested 4,000-
8,000 kilograms per hectare per year 
of marketable fish using
manures and agricultural by-products 
as feeds. Such yields are
highly profitable, yet the commercial 
production of all-male
hybrids has been sloxx to be adopted 
and provide the desired
impact on fish production in the Northeast. 
Expansion of tilapia
hybrid culture has been restricted 
by the lack of hybrid
fingerlings. This is because of several 
reasons:
1. Lack of government and private 
fingerling production
facilities.
2. Low numbers of all-male hybrid 
fingerlings produced
when compared with pure species 
fingerling production.
3. Relatively high level of technology 
needed to produce all-
male hybrid fingerlings when compared 
with producing tilapia
fingerlings of one species.
The average farmer in the Northeast 
does not have the
technical capability to produce 
tilapia hybrid fingerlings.
Extreme care is needed to maintain 
pure lines of broodstock so
that all-male hybrids can be produced 
with regularity.
Fingerling production is limited to 
government facilities that are
unable to produce large quantities 
of fingerlings at this time.
Private production of bbrid fingerlings 
will be limited, in most
cases, to large farmers with 
money to build proper hatchery
it
Pirapitinga (Cojossome bidens), top, and embequi (Coiossoma
macropomun), bottom, two fruit eaters 
native to the Amazon River
basin, have demonstrated excellent 
culture potential in experimen-
tal ponds at the Rodolpho von lhering 
fish culture center in Pente-
coste, Brazil.
facilities and hire a full-time trained 
technician to run the
hatchery. However, wealthy farmers 
have not demonstrated an
interest in investing in large-scale 
hybrid tilapia fingerling
production.
Wbile the all-male tilapia bybrid is 
an excellent culture fish
and should prove valuable in certain 
culture situations, it is the
author's opinion that the hybrid is 
not the fish to be raised by
lower and middle class farmers 
on an expanded scale
throughout the Northeast. The best 
alternatives for expanding
intensive fish culture in the Northeast 
are the culture of pure T.
nilotico with partial harvesting, an easily 
reproduced predator,
or monosex males. The average farmer 
can produce his own
fingerlings for the culture of T. nilotica, 
since the technology is
not complicated. Thus, fish farmers 
would not have to depend
on the government to produce their 
fingerlings.
Experiments with Tambaqui (Colossoma 
macropo-
mum) and Pirapitinga (Colossoma 
bidens)
Numerous Brazilian fishes have been 
tested for intensive
culture potential (10, 11). However, 
only two have demon-
strated characteristics of good culture 
fishes.
Tambaqui (Colossoma macropomum) 
and pirapitinga
(Colossoma bidens) are native to the 
Amazon River basin. Both
feed principally on fruits and seeds. 
In seasons when fruits and
seeds are not available, tambaqui 
is known to feed on
zooplankton filtered from the water. 
Both fishes grow rapidly
and are highly prized as food fish. These 
species spawn in the
Amazon River and its tributaries in 
response to rising xxaters
during the rainy season.
Lovshin, Da Silva, Fernandes, and 
Carneiro-Sobrinbo (12)
demonstrated in preliminary pond 
tests with fingerlings
captured and transported from the 
Amazon River that both
species possess excellent culture 
characteristics. However,
neither species spawned naturally in 
standing water ponds. Lack
of initial success spawning tambaqui 
and pirapitinga under
hatchery conditions limited experimentation 
due to a lack of
fingerlings. In March 1976, pirapitinga 
was artificially spawned
for the first time. In February 1977, both 
tambaqui and pirapi-
Over 8,000 kilograms per hectare per year of pir-apitinga (Colos-
soma b/dens), averaging 900 grams, have been 
harvested from ex-
perimental ponds.
tinga xxere artificially spaxxned and tarnbaqui 
xxas spaxvned
again in February 1978. IDa Silxva, Carneiro-Sobrinho, 
and F.
NMclo (1) and Lox shin (11) described this spaxxning 
plrocedure.
Both species xwere spaxxned using intramuiscular 
injections
of pituitaries taken from mature curinzata 
cant ur, Ptochilodnus
ccaren515. It wxas found that pirapihita spaxxn naturally 
after
pituitary administration wxhen the mnales and 
feiales are
placed together in a spaxvning tank. I loxxvexer, 
tambaqui muist
be stripped inf their sex prosducts to obtain xviable 
eggs. Pitapi-
tin pa mnales and females reach sexual miatuuritx 
at 2 and 3 years
of age. respectixvely . It appears that tambaqui 
mnales and femnales
reach miaturitx at Sand 4 x ears of age, 
resp~ectixvely. Preparation
of broodstock of both 'species is of utmost 
imp]ortance for
spawxning to be successful. Broodstock should he 
wxelI fed and
stocked at about 200 to) 300 per hectare. Txxo 
to 4 xx ceks before
sJpaxx-ning is anticipated, it is imiportant 
to stimiutlate both sexes
of broodstock to mnaximum Ipreparation. This 
is best accorn-
plished in areas of loxx rainfall by ix ving the 
broodstock
to a p)ond that receixves a large anmoutnt of rain 
ruin-off. This
lpond should be kept half-full sot that the first 
good rainfall
xx'ill fill it. Increase in sextual preparedness 
due to the rainwater
is imm-ediate if the fish are close to their 
spaxxning season. A
secondl method involxves mnoxing the broodstock 
to a freshlxy
filled pond. It appears imiptrtaut that the freshly 
filled pond
first he dried tor sexveral wxeeks, thereby alloxwing 
grasses tin
groxx in the pond bottomn before it is filled.
Further research should be conducted inn brood 
stock
mnanageient, s])axx ning techniques, 
and fry raising. Hoxxcxer,
annual spaxwning of both tautbagni and pita pitiutga 
ninwx app]ears
feasible.
Fingerlings of both sp~ecies fromn the 1977 slpaxxning 
season
xxere testedl in production experimients by D~a 
Silx a, Carmiciro-
Sobrinho, F. Nfelo, and Lox shin (2). A 2 x 2 factorial 
design xvas
used to test tambaqui andi pita pitiutga at txxo stincking 
densities.
Each species xxas stocked ait 5,0(10 and 10,000 pe 
hectare, each
stocking lexvel being replicated three times. Twelxve, 
355-square
mneter earthen ponds xx ith independent wxater inlets 
aud drains
xxere used. All fish we re fed a Ilelleted chicken ration 
(17 percent
plrotein) at 3 percent of the axverage standing crop of fish 
in each
treatment. Fish xxere fed in the afternoon, 6 
days a xxeek.
Between the tenth and elexenth 
mionths of the experimnent,
feeding xx as sulspended in all ponds duie to loxx dissolxvcd 
oxygen
levels. WXhen feeding xx as resummed, fish xvere fed 
at 1.5 percent
of the axverage body xxeighit of fish in each 
treatnient until
termination of the experimient. Feeding rates 
xxere recalculated
miontlxl based on sam])les of fish taken xvith 
a seine. Monthly
saitpling consisted of xweighing and mneasuring 20 percent 
of the
fish in each pond. WXater xxas added to the ponds 
only to repllace
exvap)oration and seepage losses, except when 
low dissolved
oxy gen in the tenth month required xwater renexxal 
to decrease
fertility lexvels. The experiment terminated 
after 365 days. A
summnary of the results can he found in table 6.
Statistical analysis shoxxed a significant difference 
(0).05 level)
betxxeen the axverage final xveight of tamnbaqui 
and pira pitinga
stocked at 5,000 per hectare, but no significant 
difference (0.05
lexvel) betxx een the txxo species wxhen stocked 
at 10,000 per
hectarc. Increasing the stocking rate from 5.000O 
to 10,000 per
hectare resulted in a significant decrease (0.05 
lexvel) in average
wxeight for tarnbaqui bitt no significant 
decrease in axverage
wxeight for pita pitin pa.
Axverage productions were significantly different 
(0.05 level)
betxxeen tanmbaqmxi and pita pztinga 
stocked at 5,00)0 per hectare,
bitt no significant difference (0.05 lexvel) xxas found 
betxxeen the
txxo sp~ecies xx'ben stocked at 10,000 l)er hectare. Both species
shoxxed a significant increase in axverage plroduction 
(0.05 lexvel)
xxhen the stocking rate xxas increased from 5,000 
to 10,000 per
hectare.
Wh ile tambo qui demionstrated a highem final axverage 
-w eight
antI production than pita pitiutga at both lexvels of stocking, 
only
at 5,000 fish per hectare we re these differences 
statistically
detectable. Increasing the stocking rate of tantbaqui resulted 
in a
significant increase in axverage p~rodluction, but this 
significantly
reduced the final axverage xvcighit. Increasing the stocking 
rate of
pira pitinga significanty increased axverage plroduction wxithout
significantly reducing the axverage final wxeight. 
As both species
reached harx estable size at both stocking densities 
tested, it
app)eared adx antageous to use the higher stocking 
rate.
\laxirnuimn axverage p)roduction for both tamba 
qui and
pI a pitotga wxhen stocked at 10,M(0 fingerlings 
per hectare xxas
9,391 and 8,319 kilogramns per hectare per 
year, respectixvely.
Both species reached mnarketable size in 6 mionthis, 
table 6.
Food conxversion rates at 6 mionths xxere excellent, considering
the loxx level of dietary pIrotein used. The 
12 mionth food
convxersions xx crc good, bitt these had increased because 
p)oor
wxater quality and high standing croups had loxxered 
ratiton
utilization efficiency.
Both species xxere able to xx'Phstand dissolxved 
o~xy gen levels
beloxx (1.5 mnilligram per liter for 8 ton 1(0 hioirs wxithout 
mnortality.
After 2 wxeeks of extremnelx loux nighttime dissolved 
oxy gen
lexvels, slight mortality xxvas encountered in txx 0 pita 
pitinga
TABLEu6. Stxuisins 01 l: RF:siu Lis OFsxo\ LxV\ 11-S01 .S']iadIK[\(,\
I itF (:iox ' if SN) Puowui io11 0!O FxunIA( : I (('flossouk
0 Sn tinPOMUi ) SN!) Pie SiTLNGinA (('01055Sf 5 BiMtl)t
Perfotrmttance meausure
Result. by stocking densities per ha
Tuainqfu Pmrapitiuupa
5,000 10,000 5,00)0 10,000
Ax. initial xxvt., g ... . . .25 23 
3(1 28
Results at 6 miouths
Ax. wxt. g ................. 619 424 521 
415
Ax. production/ha, kg ... 3,054 4,245 2,571 
4,154
Feed cornversion ........... 1.6:1 1.6:1 1.8:] 
1.6:1
Final resutlts'
Ax. final xxvt_ g ............ 1,496 1,052 1,064 
896
Ax. prodiatint pnd, kgz 2.37 333 
150 295
Ax. )rodiuctiou/ ha, kg . .. 6.683 9,391 4,230 
8.319
Feed conxersion ........... 2.81 2.8:1 
3.7:1 :3.0:1
Surxvixval, pet . ...... 91 87 80 
94
Growxxth, g/day ..... 4.0 2.8 
2.8 2.4
All results are the ax erage of three replicates except tanbaqui 
and
ptrapitiutga at 5,(100 per hectare xwhich are the axverage oif twxo 
replicates
due to loss sutrx ixal in one pond in each tireatment.
TABLE 7. SUMMARY OF THE RESULTS OF TAMBAQUI (COLOSSOMA MACROPOMUM) AND PIRAPITINGA (CoLOSSOMA BIDENS) IN
POLYCULTURE WITH THE ALL-MALE TILAPIA HYBRID (FEMALE TILAPIA NILOTICA X MALE TILAPIA HORNORUM)
Result, by stocking densities
Performance measure Tambaqui + all-male hybrid Pirapitinga + all-male hybrid
5,000/ha 5,000/ha 5,000/ha 5,000/ha
Av. initial wt., g ............................ .. 25 18 33 20
Results at 6 months
Av. wt., g ......... ................ 485 245 383 261
Av. production/ha, kg .. ............... 2,393 1,209 1,890 1,288
Feed conversion ............................ . 1.7:1 1.2:1' 1.8:1 1.0:1
Final results
2
Av. final wt., g ............................. 1,189 748 1,045 681
Av. production/pond, kg .................... 200 117 170 128
Av. production/ha, kg ...................... 5,640 (8,939) 3,299 4,794 (8,404) 3,610
Feed conversion ............................ 2.8:1 1.8:1' 2.7:1 1.8:1
Survival, pct................................ 95 89 93 99
Growth, g/day ............................. 3.2 2.0 2.8 1.8
'Feed conversion is calculated using the total weight of Colossoma and tilapia hybrids.
2
Results are the average of three replicates.
ponds. No mortality due to dissolved oxygen was encountered
in ponds with tambaqui. The poor water quality slowed the
growth of both species, and only after dissolved oxygen levels
were corrected did normal growth resume.
A second experiment involving the polyculture of tambaqui
and pirapitinga with all-male tilapia hybrids was run
concurrently with the preceding experiment. A completely
random design was used with two treatments replicated three
times each. Tambaqui and pirapitinga were stocked in 355-
square meter earthen ponds at 5,000 per hectare along with 5,000
all-male tilapia hybrids per hectare (combined stocking rate of
10,000 fish per hectare). Fish were fed 3 percent of the average
body weight of tambaqui and pirapitinga only. Fish received a
pelleted chicken ration (17 percent protein) in the afternoon, 6
days a week. Between the tenth and eleventh months of the
experiment, feeding was suspended in all ponds due to low
dissolved oxygen levels. When feeding was resumed, fish were
fed at the rate of 1.5 percent of the average body weight of
tambaqui and pirapitinga until termination of the experiment.
Feeding rates were recalculated monthly based on seine
samples. Monthly sampling consisted of individually weighing
and measuring 20 percent of the fish populations in each pond.
After 365 days, ponds were harvested. A summary of the results
can be found in table 7.
Analysis of variance (0.5 level) demonstrated no significant
difference in the final average weights of tambaqui and
pirapitinga. However, there was a significant difference (0.05
level) in the final average productions. Again, tambaqui had a
higher production, 5,640 kilograms per hectare, than
pirapitinga, 4,794 kilograms per hectare. Tambaqui and
pirapitinga grown in polycultures with the all-male tilapia
hybrid had total productions approximately equal to production
of both Colossoma species in monoculture at 10,000 per hectare.
The use of tilapia increased the total productions of the
polycultures by more than 3,000 kilograms per hectare. The
added production of tilapias was realized without an increase in
the feed used. Feed conversions demonstrated that both species
of Colossoma were utilizing the feed efficiently. Conversion
efficiency was equal to or better than the monoculture of these
species at 5,000 per hectare, tables 6 and 7. Apparently, the
tilapia hybrids were produced on planktonic algae and organic
wastes generated by the tambaqui and pirapitinga. Thus, the
polycultures are much more efficient and economical than the
monocultures, resulting in higher productions with less feed.
Both tambaqui and pirapitinga have demonstrated
outstanding culture characteristics. Both species are resistant to
handling, poor water quality, and diseases. They feed on a wide
variety of fruits, seeds, agricultural by-products, and formulated
rations, and are extremely easy to capture with a seine. Although
tambaqui \and pirapitinga have good quality flesh, large
intermuscular bones may limit consumer acceptance. Both
Colossoma species have been artificially reproduced. These
two tropical, freshwater species have shown more fish culture
potential than any other South American fishes tested to date.
The results discussed here support the exciting possibility that
these fish can be cultured effectively. Further work with the
species should be pursued vigorously.
RESERVOIR FISHERIES
Since 1911, DNOCS has constructed 254 public reservoirs and
aided in the construction of thousands of small private reservoirs
in the Northeast. These reservoirs collect and retain run-off
water during the short rainy season, providing water for
irrigation, livestock, and human needs during the dry season. To
better utilize these waters, the Department of Fisheries and Fish
Culture within DNOCS began a fish research and stocking
program in 1932 and 1942, respectively, to increase natural fish
production in these reservoirs.
It soon became evident that the limited fish fauna of the semi-
arid Northeast was not sufficient in species diversity to
significantly increase reservoir production. Thus, DNOCS
biologists began testing and stocking species of fish from river
basins outside the Northeast and exotic to Brazil. Fish were
selected for their economic value, ability to reproduce in
reservoirs during periods of low rainfall, and ability to fill an
underutilized ecological niche. Two species of fish were
introduced from the Sao Francisco River, one species from the
Parnaiba River, five species from the Amazon River, and Tilapia
rendalli and nilotica from Africa.
DNOCS controls the commercial fishery in 103 reservoirs (6).
In 1977, a total of 14,788 metric tons of fish and shrimp was
captured from these reservoirs. Of this total, 85 percent was
composed of seven species, of which five were introduced to the
Northeast, table 8. The 1978 catch totaled 19,478 metric tons of
fish and shrimp. The same seven species composed 90 percent of
the total fish and shrimp catch, table 8.
Tilapias have proven to be an excellent addition to DNOCS
reservoirs. From 1970 through 1978, total production of T.
rendalli and T. nilotica from DNOCS controlled reservoirs was
5,859 and 7,332 metric tons, respectively. This represented 13
percent of the 101,520 metric tons of fish captured over the 9-
year period. Introduced to the Northeast in 1960, the
herbivorous Tilapia rendalli was the dominant tilapia species
captured in DNOCS reservoirs until 1977. In 1977, however,
T. nilotica surpassed T. rendalli in kilograms captured even
though T. nilotica, introduced to the Northeast in 1971, is found
in fewer reservoirs than T. rendalli. Tilapia nilotica first entered
11
TABLEcS. PI-IN(IPAIN J\xs Exo\)'. 
I J
N" 03> ID %C\s omo...I 
1
Common
name
Pescada do Piaui
Canela shrimp
Peacock bass
Traira
Nile tilapia
Curimata comun
Congo tilapia
Others
Scienic
name
Plagioscion squamosis.i
Mfacrobrachium amazonicu.
Cichla ocellris
Hoplias malabaricus
Tilapia nilotica
Prochilodus cearensis
Tilapia rendalli
.1.5i~:
'!' ( , 5 '-'I151
'v
Tiapia riloca, widely culured in 3ords 
o0& B C
Northeast Brazil, is a highly regarded food 
fish. -- er 4,0 TrIcC
tons of this species were captured in DNOCS controlled 
eservoirs
in 1978.
the capture statistics in 1974 when 16.1 metric 
tons w ere cauglht.
In 1978, ten reservoirs produced 
4,842 metric tons of T. 
nilotica,
making this species the principal fish captured 
in DNOCS
reservoirs just 5years after its introduction. 
It is expected thla the
production of T. nilotica will continue 
to increase as it ecComes
more widely stocked.
Lovshin, Peixoto, and de Vasconcelos 
(15) demonstrated t'he
economic and fish production benefits of 
stociCng tilapias into
reservoirs in the Northeast. Tilapias 
adaps weli to most
reservoirs, reproduce independent of the 
rainy season, are easily
captured with cast nets and gill nets, and are 
readily accepied h
consumers at a good price. Stunting of tilapia 
populations in the
reservoirs has not been a problem.
Stocking of fish species not native to Northeast Brazil into the
many public and private reservoirs 
located in this region has
proven highly successful. Fish captured 
from these Interior
waters are almost entirely consumed 
in rural areas wvhere
reservoir fish are priced lower than 
other meats. Over 10,000
licensed fishermen gain at least a part 
of their yearly income
fishing in DNOCS controlled reservoirs. 
The Brazilian
experience in reservoir 
construction, fingerling 
fish production,
stocking, research, and regulated 
capture should serve as a
model for the whole tropical world.
. 118T)
p - .
- C
1xiti~i 9. Co -lit mivn NI ill 'Nsxix o, I li i NxI\ %I
li llmilo r\ x 0..3544:i: I il Po\
1. Costs.
Fixwd costs'
Pond,' lni nac .....
CrIF.ilritlli' 1:92k
a('r.$1 2111)........
Nt iti'llilii'.............
'r.83,5101)0
Cr.8 1001.00)
l'lntl consitructionl xxas pbaid] xwithi probtits fro ii(t'e fir
'81.X 0(1S. -8 Cr81211).
fariiier built a .3,500 xtjlart' ineter fariii lptii
rlelresxillni belhxx en lR1holding 12(1 younig calxvex.
stoetd xxith 4,000 tilapia hy bridsx axveraging 48 g
wxaxte's fromn the calve r'xxxerr' xxaxhed illo thr' poll
o~ther fertiliz'er xxax used. AXfte'r 1.31) dax s, ther pint(
and 1 :392 killlgranis (:3.!77 kilograin'.per herctar'
axverage xx\,'ighit o~f 40(1 grains xxert' harv exte'd. PrI
day culture pheriodt xx ax Cr. $12,.514.(X) ($1,015),
T[wo eIrarthen polnds oIf 2,301 square mietters ('at
the IIN C S \ llradta Notixa irrigat ion pIrojet t bra
ohf Ceara. These' pontdx xxverr' dug inl xaliner xoils tha
tradtitio~nal aigricultural trollpxand reeixved xxater1
frtn i irrigation canial. Fath pond receixetd
tilaiiia hy bridx xxith ant axverage' xxeight o~f 13 grali
1)NOCS hatche'ries free of charge. Fixh xx ere fe'd
at :3 petrc'ent of their bodry xxeight fin eac'h lVoi
fertilized xxithi 1.54 kilolgramsi hper xx cek (6370
hie'ctare') of fre'xh cattle' inanilre. Hit't b~ran antI
xxre by-p~rodts~t prodtuced't inl ther irrigatioln I~rl
tdaysx, hitli pondis~ xxere harx exted. Ploid 110
killlgranx o~f fixh (2,917 killlgram'. pe'r hec'tare') x'
xx eight of 2.35 gramxi aiid a food conxverxionu (If 4.11
Ci II I ii 11 xield ed 917 k iiogra ii oIf tilai a hy bridsx (.3,983 k-ill graini 1er
ii ii (~ecta re') xx ithI ani axIrag' \\elight o f 384 gra ins antd a food )t
lit converstio xofil~ 2.9 to 1. Proffit for both 
pionds ox er thle 180 clax
ciilture periotd xwas Cr. $:3,502.25 (S250), table 10.
I \N oI taithel ('0 ld o11( f 5,0(00 sua re iiiete rx eachtl x ere built ill
tlit e I) N ( (' P enit ('( Ite irri gatio ilp ro ject . The p oflds wxere
Cr$3I1A located 
oil salIinet l andi uinf 
it fo(r iri igated ro Ips. 
Po ndl one'
Cr 8: (31 IX) recefix'd .5.000 tilapia lix riti fingerlings wxith ant axv(rage wxeight
oIf 1 Igira ins PiN111lxx iN o receixved 5(,6)0 tilahi hyixbrid fin geriiings
ax eragig 18 grinx. Finigerlinigs wxere r'ov)Xided' bx the
('r.$ 480.0 1) entt'rott' fish cult) ire rese(archi statioln free of ehlirgr. Fish xx ere
Cr$4,090.0 fed1 h tIxxliat I rai i at 31 pe'rcent 
o f tI eir hiody xxeight tlai lx . Eacrh
Ihlt xwasi fertilized xwith 1,0)00( killlgrain'. of cattle' iiianlire 2
('r.816,704.01) wxeeks llefol e stockinig anid rteceixved .5(0( killhgrailnx perr xxeek o~f
C'r.$12,1614.001 fresht rattle iiianiire aifter stoceking. 'I'lit' cattle inanlire xx as a bx -
IrS s SI'm (.5 1-,) producit lof a dfairx, ope~ratioln 
located iii the( irrigation prIojr't.
st cultu re periodIt Al lei 2.5:3 dIaxs poI ntl one i'x ax ha rxvexsted by the fanrimers inl the
iriigatio cli 111leralix e' rt'xtltfing fin a harxve't o f 1.957 kih Igra insi
(:3,914 kiliigiam pie~ lr hiectart') (If fixh xx th ant axv(rage xweight o~f
d1 inl a natuiiral 40)2 gr a ins . Foold con iiioxt x n xxax 2.5 toI 1. TIhe seco nd pn id,
'I'he 11111( wxax liarx estr'd lbx tdraininig aftr'r 229 daxyx, yir'lder'd 2,249 
kilolgranms
rains etat'hi. 'Ih e (4.498 kih Igra iii xper' herctare') o f til apiai i hbrids ax rragin 
g 4:30
dI tailx, .and no grainx. Th le iate' of food( ('liuxerxiln wxax 2.3 toI 
1. Fish fromi both
Ix was harxveste'd pontds xxer I'll' d~ onl the pollt hank toI a supellrmiarke't chainl inl
) of fisb wxithill Flurtaleza folr Cr. $ 1:3.50 ($0.8,5) peri killlgrabn. Profit for both
)fit folr ti'( 1:31) p ids \l'\x ax Cr. $21),522.7.5 ($1,28:3), table' 
11.
table' 9.
If xxere'buiilt ill xifll 10.I ('oxilli1i 
osx 13JI NI ill HxAI lO lilF IVI FASxlxF Cl 1,11t1R
IIIHi lix )N\ li umiio N lxo1:2.11 Iro PXoIII 1'IIN~ I I
t xx t're' unfit fo r
IN griix itx flllx
2,41( all iiiale
is, prllxith'lI bx
rice bran tdailx
d. Pndx xx e're
tattr' inre
Ijr'et. A ftt'r 180
' xyielded 6371
vitli an axverage
t1. 1PFlnd txxoI
:
Iteni)
1. C'osts
G liil tI Set-i ( .. ... ...
(10) x,'.rs...............
'l'ot ,i tix l I I'll')'. s.........
IlI(ic bran (5388 kg at
Cttl mamir) k 800g k)
I Ii x t labour ..........
G:ross initcomel (1,.587 kg
at ('.SI0.75) .........
Ne t illcoIii'..............
Xnioulnt
0'i.$164.)00
C'r.$7,(079.100
Cr.84,310) 00(
Cr.$1 205)111
Cr8$7,84300X
Cr8$ 5,715.00)
Cr. $13,558.1(0
(:r.$17 (161.25
('r.S 3,502.25
(1. 8. $250.010)
'81.) t. -. Cr8$14 1)1)
'I xlii 1l.('xIl~h'x l Ai[iiil 13i 1111 ii INI I x\ E Ci tLTlilR
III Tii xix lhxlimx i\ 'l'xx, ((.5111(1~ ui. Pom'xo I\li
Itemii
Guard tIXln itt'............
10) x cars)..............
'ota i\i cxtsts t.........
V Variaileh cI 
b..
Mict ib(ran (9,7.55 kg
iatle (nnr r 3,0 kg)' g
ToI , I, l o1 I'..............
2.I lil '.
Grs i iconin (4,20)6 kg
a r.$1:3.5) . .. . .
NOilcnut.......
Amiiounilt
Cr.$ 1,6151)1
(:r.$15. 46) IX)
0i.$ 8100X
Cr.89,532.25
Cr. S8,8501.10)
Tilapia raised in DNOCS irrigation projects are being sold in super
markets in Fortaleza, Brazil.
Cr.$17,876.01)
Cr.$ 18,'382.25
Cr.$ 36,2.58.25
Cr8$ 56,781.00
Cr8$ 21)522.75
(1.S. 81,282.67)
I
$LM U.S. - Cr.$16A).

BIBLIOGRAPHY
(1) DA SILVA, A. B., A. CARNEIRO-SOBRINHO, AND F. R. MELO.
1977. Desova Induzida de Tambaqui, Colossoma macro-
pomum, Com o Uso de Hipofise de Curimata comum,
Prochilodus cearensis. I. Symposium of the Latin Ameri-
can Aquaculture Assoc., Maracay, Venezuela.
(2)
AND L. L. LOVSHIN. 1978. Mono e Policultivo Intensivo do
Tambaqui, Colossoma macropomum, e do Pirapitinga,
Colossoma bidens, corn o Hibrido Macho dos Tilapias,
Sarotherodon niloticus female e Sarotherodon hornorum
Male. Second Symposium of the Latin American Aqua-
culture Assoc., Mexico City, Mexico.
(3) 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.
(4) DE CARVALHO, J. N. AND J. A. FERNANDES. 1978. Criacao
Intensiva de Peixes em Perimetros de Irrigacao do
DNOCS. Second Regional Directory, DNOCS, Forta-
leza, Brazil.
(5) , AND J. A. DE
OLIVEIRA. 1977. Criacao Consorciada de Hibridos Oe
Tilapia Zanzibar, macho Sarotherodon hornorum x
Tilapia do Nilo, femea Sarotherodon niloticus e Bovinos.
Second Regional Directory, DNOCS, Fortaleza, Brazil.
(6) DE VASCONCELOS, E. A. 1970-1978. Quadros Informativos
Sobre a Administracao da Pesca em Acudes Publicas
Controlados Pelo DNOCS. Central Administration,
DNOCS, Fortaleza, Brazil.
(7) JEFFREY, N. B. 1974. Progress Report on Fisheries De-
velopment in Brazil. Project AID/csd-2270, Task Order 3,
International Center for Aquaculture, Auburn Univ.,
Auburn, Ala.
(8) 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.
(9) . 1976. Progress Report on Fisheries De-
velopment in Brazil. Project AID-1152, Task Order 2,
International Center for Aquaculture, Auburn Univ.,
Auburn, Ala.
(10) LOVSHIN, L. L. 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.
(11) . 1977. Progress Report on Fisheries De-
velopment in Northeast Brazil. Project 
AID/csd-1152,
Task Order 2, International Center for Aquaculture,
Auburn Univ., Auburn, Ala.
(12) , A. B. DA SILVA, J. A. FERNANDES, AND A.
CARNEIRO-SOBRINHO. 1974. Preliminary Pond Culture
Tests of Pirapitinga (Colossoma bidens) and Tambaqui
(Colossoma macropomum) from the Amazon River
Basin. FAO Aquaculture Conference for Latin America.
Montevideo, Uruguay.
(13)
. 1974. The Intensive Culture of the
All-Male Hybrid of Tilapia hornorum male x Tilapia
nilotica female in Northeast Brazil. FAO Aquaculture
Conference for Latin America. Montevideo, Uruguay.
(14) .1975. Culture of
Monosex and Hybrid Tilapias. FAO/CIFA Symposium
on Aquaculture in Africa. Accra, Ghana.
(15) , J. T. PEIXOTO, AND E. A. DE VASCON-
CELOS. 1975. Tilapia sp. in the Northeast of Brazil: Eco-
nomic and Ecological Considerations. Symposium of
Limnology, Fisheries and Fishculture, Belo 
Horizonte,
Brazil.
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 os
Secas (DNOCS)
Av. Duque de Caixias, 1700
Caixa Postal 650
60.000-Fortaleza, Ceara, Brazil
15