- ft
-4ISO.,
........
. .. 0. ..
International Center for Aquaculture
Alabama Agricultural Experiment Station, Auburn University
Lowell T Frobish, Director Auburn University, Alabama
Research and Development Series No. 35 September 1990
TABLE OF CONTENTS
Page
IN TR O D U C T IO N ........... ......................................................... 3
PRODUCTION OF FRY FOR SEX REVERSAL ......................................... 3
C O NCEPT .......................................................................... 3
INPUT REQUIREMENTS..............................................................3
PRO CED U RE ....................................................................... 4
EXPECTED RESULTS ................................................................ 6
PREPARATION AND ADMINISTRATION OF HORMONE TREATED FEED.............7
C O NCEPT ........................................................................... 7
INPUT REQUIREMENTS..............................................................7
PROCEDURE ........................................................................ 8
EXPECTED RESULTS ................................................................ 10
COMMON QUESTIONS ON THE PROCEDURE ....................................... 11
SUGGESTED READING ON SEX REVERSAL .......................................... 
14
FIRST PRINTING 3M, SEPTEMBER 1990
PREFACE
This Aquacultural Production Manual from the International Center for Aquaculture at Auburn
University is oriented towards field application rather than rigorous scientific evaluation of the de-
scribed techniques. The management practices have been field tested and proven feasible at the
described scale of production.
The intended readers are intermediate to upper level technicians with experience in related
aquacultural activities. As such, a basic understanding of the principles of aquatic production and
of the life history and environmental requirements of tilapia is presumed.
ACKNOWLEDGMENT
Complete cost of publishing and partial salary support
of the authors were provided by the United States
Agency for International Development through a Co-
operative Agreement and Program Support Grant
to Auburn University.
Sex Reversal of Tilapia
in Earthen Ponds
Thomas J. Popma and Bartholomew W. Green,
INTRODUCTION
ON COMMERCIAL fish farms female tilapia are usually ex-
cluded from growout ponds to prevent overcrowding and stunting
resulting from unwanted reproduction. Females can be eliminated
manually by visual inspection of the urogenital papilla of juvenile
fish, but this technique is labor-intensive. The production of all-
male hybrids is also technically feasible, but isolation facilities to
maintain the purity of both parental lines are essential, and seed
output is often reduced due to partial reproductive incompatibility
between two different species. A more recent technique for pro-
ducing male fingerlings is "sex reversal" or "sex inversion." This can
be achieved by offering feed treated with a male hormone to tilapia
fry before the primal gonadal cells of females have differentiated into
ovarian tissue. The dietary hormones can be suspended once the
testes are sufficiently developed to maintain normal levels of endog-
enous hormone.
Until the early 1980's, sex reversal of tilapia was conducted in
clear water to reduce possible interference from natural foods. The
water source was generally a well or dechlorinated and oxygenated
municipal water. Treatment was usually conducted in shaded con-
tainers to retard the development of phytoplankton. The quantity of
feed administered required that the containers be cleaned daily and
that water quality be maintained by frequent water exchange and
aeration. Under these conditions, heavy losses from disease and par-
asites were common as the technology shifted from experimental to
commercial levels. The presumed requirement of clear water did
not render sex reversal technically or economically unfeasible for ef-
ficiently run operations, but it greatly limited application in small
and medium scale fish farms, especially in developing countries.
Since 1984, investigators and producers in several countries, in-
cluding the United States, Israel, Thailand, Philippines, Jamaica,
Ecuador, Panama, and Honduras, have demonstrated that tilapia
can be effectively sex reversed on a commercial scale in hapas (fine
mesh net enclosures) held in ponds with abundant natural plankton.
The fish grew faster, the incidence of disease was reduced, the basal
diet did not have to be nutritionally complete, and the infrastructure
requirements were reduced. The procedure, as described in this
publication, is relatively simple and can be practiced by producers
without wells, pumps, or indoor wet lab facilities.
The techniques and results described in this publication are
based on experiences by the authors since 1986 at the Escuela Su-
perior Politecnica del Litoral (ESPOL) in Guayaquil, Ecuador, and
at the El Carao Aquacultural Experiment Station in Comayagua,
Honduras, with Nile tilapia, Oreochromis (Tilapia) niloticus. The
size of the production units described are the same as those used for
field operations. Extrapolation to somewhat smaller or larger units
is probably valid, but upscaling would usually be accomplished by
'Popma is Associate Professor and Green is Senior Research Associate in the De-
partment of Fisheries and Allied Aquacultures at Auburn University.
increasing the number of production units rather than the size of the
unit itself.
This publication is organized to facilitate field application of the
hormonal sex reversal technique for production of all-male finger-
lings of tilapia. Detailed discussions on the justification and accept-
able modifications or alternatives are minimized in the main body of
the text. A final section, entitled "Common Questions on the Pro-
cedure," is included to assist managers in decisions on the general
appropriateness of sex reversal, on site specific modifications, and on
overall farm management.
PRODUCTION OF FRY FOR SEX REVERSAL
Concept
The technique to produce fry for subsequent sex reversal is based
on the following major considerations. (1) Sex reversal of tilapia must
begin before the gonadal tissue of young genetic females has differ-
entiated into ovaries. (2) Depending on water temperature, large
numbers of free-swimming fry can first be found in warmwater
ponds about 2 to 3 weeks after stocking healthy brood fish, but soon
afterwards the number decreases as small fingerlings become can-
nibalistic on the recently hatched fry. (3) Fry production must be co-
ordinated with the sex reversal operation because the hormone
treatment should be begun immediately after fry harvest and the fa-
cilities will not be available to receive the following batch of fry for
another 25 to 30 days. (4) For efficient farm management, it is gen-
erally preferable to sex reverse infrequent batches of large numbers
of fry rather than more frequent batches of fewer fry.
Based on the above considerations, the suggested procedure for
fry production in ponds involves a 25- to 30-day cycle (including a 2-
to 10-day turnaround period) with a single, complete harvest of fry
not exceeding 14 mm total length. For most efficient use of labor and
facilities, fry production may be accomplished in a single pond.
However, additional ponds may be required 
depending on the num-
ber and size of ponds available and the volume and frequency of de-
mand for sex-reversed fingerlings.
Input Requirements
(1) Earthen pond with harvest basin: The results described were
obtained from 500- to 2,000-m
2 
ponds, but extrapolation to smaller
and somewhat larger ponds is presumed valid. An earthen harvest
basin should be 30 to 40 cm deep 
with a surface area of about 10 m
2
or 1% of the pond area, whichever is larger. Concrete harvest basins
may be somewhat smaller and also make fry collection more conve-
nient and maintenance less labor intensive. Complete harvest of fry
is not recommended in a pond without a harvest basin.
(2) Water source with salinity less than 10 parts per thousand
(ppt).
(3) Nylon netting (13- to 20-mm or 1/2- to 3/4-inch square mesh)
with dimensions that exceed those of the harvest basin by 20%.
(4) Fine mesh (1.6-mm or V6-inch) hand-held scoop net.
(5) Fine mesh filter screens for pond draining.
(6) Hapa or cage for temporary holding of brood fish between
cycles (3- to 4-in
3 
capacity per 100 kg of brood fish).
(7) Brood fish (70 to 100 kg for production of 50,000 target-size fry
per cycle), sex ratio of 1.5 to 2 females per male.
(8) Supplemental feed (40-50 kg of feed per 100 kg of brood fish
per cycle).
(9) Fish grader, as described in Step 10 of Procedure.
(10) Labor requirements per cycle in a 1,000-m
2 
reproduction pond
stocked with 100 to 150 kg of brood fish: 15 person-hours for pond
preparation, stocking of brood fish, and feeding, plus 15 person-
hours, divided among 3 to 4 persons, for pond draining, harvesting
and grading of fry (For 500-m2 reproduction ponds, the labor re-
quirements decrease by 30%; for 2,000-inm
2 
ponds the labor needs in-
crease by 50%).
Additional details are provided in the Procedure section.
Procedure
The procedure to produce fry for subsequent sex reversal is de-
scribed in 10 steps:
(1) VERIFY WATER SUPPLY CHARACTERISTICS
Chemical composition is adequate for fish growth and average
water temperature in ponds is at least 22
0
C, but not more than 32
0
C.
The salinity of water must be less than 10 ppt and, ensure the quan-
tity of water is sufficient to drain and refill the pond each cycle.
(2) ELIMINATE ABSOLUTELY ALL FISH IN POND BE-
FORE FILLING
This can be accomplished by complete drying or by disinfecting
puddles with rotenone or swimming pool type granular chlorine
(generally calcium hypochlorite) dissolved in water. In either case,
apply 10 to 20 g per m
3 
of puddled water. (This dosage is based on
the assumption that the rotenone is 5% active ingredient and the
commercial granular chlorine is approximately 60% active.) The fol-
lowing day verify the complete absence of fish because some small
puddles may have been missed. Repeat the treatment if live fish are
found.
Another effective technique which is less labor intensive but more
costly, is to disinfect with rotenone after filling the pond with just
enough water to cover the pond bottom. If average water depth is 10-
20 cm, approximately 750 ml or g of commercial rotenone will be
needed in a 1,000-m
2
' pond. Ponds should be drained to verify the
complete absence of fish before restocking. To prevent downstream
fish kills after draining, either wait 1 or 2 weeks for natural degra-
dation of the rotenone or apply 2 to 3 g of potassium permanganate
(KMnO
4
) per ml of 5% rotenone for immediate neutralization of the
rotenone.
Quicklime (or burnt or builder's lime) has given less consistent re-
sults, especially in ponds with organic-rich bottom muds. Approxi-
mately 100 kg per 1,000 m' surface area of pond bottom should be
applied if the more effective fish irradicants are not available.
(3) DRAPE THE LARGE MESH NETTING IN THE HAR-
VEST BASIN, FIGURE 1
Secure at the borders so that few brood fish will swim under the
netting.
(4) FILL THE POND WITH WATER SO THAT WATER
DEPTH IN MOST OF THE POND IS 50 TO 80 CM
If necessary, filter water to prevent entry of wild fish. In some re-
gions, manuring may be required to control excessive clay turbidity,
but fertilization to stimulate plankton growth is not normally rec-
ommended. Plankton production in a previously fed pond is usually
adequate to maintain nutritional health of the fry
(5) STOCK ALL BROOD FISH ON SAME DAY, SOON
AFTER POND HAS BEEN FILLED
Fry production is more a function of brood biomass than of pond
size. To produce 50,000 acceptable size fry for sex reversal, stock 35
to 55 kg of mature females and sufficient males so the sex ratio is 1
male per 1.5 to 2 females. The pond area required to safely and ef-
ficiently accommodate this biomass 
is 500 to 1,000 m
2
'.
(6) MANAGE POND DURING SPAWNING/PRIMARY
NURSERY PERIOD
A supplemental feed is recommended, but do not overfeed. Most
commercial fish rations at a daily rate of 1% of fish biomass will main-
tain brood fish in good spawning condition. A slightly higher daily
feed rate (2-3% of fish weight) is recommended only when feeds are
of lower quality or the brood fish are undersize. Low dissolved ox-
ygen is seldom a serious problem. Predators, especially birds, may
have to be controlled in some regions. Frog and toad egg masses
should be removed from the pond because sorting tadpoles from fry
at harvest may produce high fry mortality.
(7) DRAIN THE POND 14 TO 23 DAYS AFTER STOCKING
BROOD FISH
The primary factor determining the appropriate harvest date is
water temperature. A suggested guide is:
Average water temperature Cycle duration
less than 25'C 20 to 23 days
FIG. 1. In fry production ponds the placement of net over harvest sump bottom is critical.
[4]
.. 30 cm
Harvest sump
1
25 to 28
0
C
more than 28
0
C
17 to 21 days
14 to 18 days
In planning the harvest of fry, consider the following factors:
The pond should be drained slowly to prevent excessive loss of fry
when stranded on the pond bottom.
Fry should be harvested in the early morning before increasing
water temperature endangers their survival.
Dissolved oxygen levels should be at least 4 mg/1.
During drainage, the water must be filtered through a fine mesh
screen (such as window screen) to reduce fry losses. When 
the sur-
face area of the filter screen is increased, the velocity 
of the drain
water and the number of fry being impaled on the screen 
are re-
duced, figure 2.
As a general "rule of thumb," the following maximum mesh 
size,
minimum area of the filter screen and total drainage time are 
rec-
ommended:
Pond size
Maximum mesh size (mm)
Area of filter screen 
(m2)
Drainage time (hours)
500 m
2
1.6
0.5 to 0.8
5 to 10
2,000 m
2
1.6
1.0 to 1.5
10 to 15
(8) AFTER POND HAS DRAINED, LEAVING WATER ONLY
IN THE HARVEST BASIN, REMOVE BROOD FISH BY LIFT-
ING COARSE MESH NETTING ON THE BOTTOM OF THE
HARVEST BASIN
These same fish may be used as brooders in the following pro-
duction cycle, which may begin in 3 to 10 days depending on the an-
ticipated availability date for the sex-reversal facilities. Temporary
holding for easy recapture and restocking is therefore desirable 
for
farms that anticipate a continuous program of fry production. 
If a nu-
tritionally balanced diet containing mineral and vitamin premixes
and some protein of animal origin is available, the fish 
may be tem-
porarily kept at low densities in holding cages (about 30 kg/ 
m3), and
fed a high quality feed at a rate of 2 to 3% daily of body weight.
It may be more productive to keep the sexes separated during this
turnaround period, but the results described in this publication
were obtained by maintaining all brood fish in a single cage sus-
pended off the bottom of the pond or in concrete tanks. If males and
females are held together in concrete tanks or in structures which
allow the females to spawn and pick up the released eggs, it is ap-
propriate to examine the females' mouths and remove all eggs before
restocking the spawning pond.
If a high quality supplemental feed is not available, brood fish that
must be maintained for more than a week between cycles should be
released in an enriched pond.
(9) COLLECT FRY FROM HARVEST BASIN SOON AFTER
BROOD FISH HAVE BEEN REMOVED
Water exchange in the harvest basin is usually not required as 
the
oxygen demand is greatly reduced once the adult fish have been har-
vested. On the contrary, the turbidity that often accompanies 
water
exchange may weaken the fry and make collection more difficult.
Most fry in the harvest basin will soon come to the 
surface if not
disturbed. They may then be harvested by skimming 
with a 40- to
100-cm-wide scoop net made of materials such as 1.6-mm 
mesh ny-
lon netting or fiber glass window screen.
Fry are delicate at this size, so special care is required to reduce
stress from turbidity and handling.
Collect first around the edges of the harvest basin; this 
is espe-
cially appropriate for larger basins that are not concrete lined.
Take special care to keep the scoop nets off the bottom muds when
collecting fry.
Transfer the fry quickly to fresh water; never keep them out 
of
water more than about 20 to 30 seconds.
Wash the fry off the scoop nets from the back side of the netting
into a holding bucket rather than shaking or pushing them off the
net.
Never keep the fry more than 10 minutes in the holding 
bucket
before transferring them into clear, well-oxygenated water.
Fry collection from the basin is a relatively fast operation. Two
persons can harvest 50,000 to 100,000 fry in less than 
an hour.
In well-drained ponds, the collection of fry that were stranded 
in
puddles is not recommended since the labor requirements are 
high
and the number and subsequent survival of these fish are low.
Large numbers of air-breathing insects are often collected 
along
with the fry. When the entire harvest is placed in still water for a 
few
minutes, the two groups often stratify vertically, with the fry 
re-
maining deep and the insects near the surface. Most of these 
insects
can then be removed by skimming. Alternatively, an oil:diesel 
fuel
mixture (1:10) can be applied at a rate of 2 gallons per 
1,000 m
2 
prior
to harvest, but this practice is not recommended because 
of equip-
ment fouling 
and possible deleterious 
effects on the fry./
(10) IMMEDIATELY GRADE HARVESTED FRY TO 
RE-
MOVE THOSE INDIVIDUALS LARGER THAN 14 MM
Graders with the required characteristics are not commercially
available but are easily constructed. The following considerations
FIG. 2. The effective filter area for standpipe type drains is increased by a 
rigid oversize wire bonnet that fits over the standpipe; the fine mesh
screen material then fits over the bonnet.
[5]
Heavy gauge
wire frame
Fine mesh
/ filter sock
the! t~llnwlnP nr~-
are suggested:
An in-pond technique for temporary storage and continuous 
grad-
ing of several batches of fry is a floating grader placed within a 
fine
mnesh hapa (1.6-mm or less). The 
volume of the net enclosure should
be at least 20 times greater than the interior volume of the floating
grader to stimulate movement of fry from the grader into 
the net en-
closure.
3.2-mm (1Vs-inch) metallic screen or "vexar" is acceptable 
material
for grading. "Target size" fry pass through the screen 
and are ac-
ceptable for sex reversal; "oversize" fry are retained 
in the grader
and subsequently rejected, figure 3.
The selectivity of a new grader should be tested because 
the
gauge of the screen material and the coating process 
influence the
size of the openings. The mesh selected should be small 
enough to
retain all the 16-mm fry and 'Most of the 15-mm fry, but large enough
for most 13-mm fry to pass. If more than 10 to 15% of the 13-mm 
fry
are retained, a larger mesh screen should be tested. If more 
than 5
to 10% of the 15-mm fry pass through the grader, a uniform 
coat of
paint may be sprayed on the screen to reduce the size 
of the mesh.
A grader with 1 m
2 
of effective surface area is sufficient to grade
at least 50,000 fry. Smaller graders also can be used, but the capacity
is proportionally reduced.
The size of a floating grader can be decreased and its efficiency
increased by bending the screen in the bottom of the grader 
to form
a corrugated grading surface. The corrugation orients 
the mesh
more perpendicular to the normal swimming position of the 
fry, fig-
ure 4.
The length and width of the grader depend on the degree of cor-
rugation and the required capacity of the grader. To process 100,000
fry over a 30-minute period, the floating grader need be no larger
than 90 cm long by 70 cm wide.
Sr
Fine mesh hapa arget size fry
F1 1 111111Grad er u s edlt l t o rIII i 
ttIIJI/isI1 11%t t f y
FIG. 3. Grader used to reject over-sized fry.
Fry are transferred either directly from the harvest basin or from
a holding tank to the floating grader. After about 15 minutes the
grader is raised and lowered several times to excite the remaining fry
to escape through the mesh. The objective is to crowd the fry with-
out lifting them completely from the water. The grader is then re-
moved. Fry remaining in the hapa are of an acceptable size for sex
reversal. Oversize fry retained in the grader are not acceptable for
sex reversal, but some can be reared for brood stock replacement.
Expected Results
Under the management conditions described, figure 5, average
results per spawning cycle from 50 kg of female brood fish (plus a
male for every 1.5 to 2.0 females) in 500 to 1,000 m
2 
of pond space
are:
Total number of fry harvested:
Target size fry (14 mm or less)
Percent of total:
Total number:
Harvest/handling mortality:
60,000 to 80,000
60 to 100 %
40,000 to 60,000
5 to 15 %
The most common causes of failure to achieve expected results
from an individual production cycle include the following:
(1) Inappropriate harvest dates. Spawning and egg/larval devel-
opment is temperature related. See Step 7 of Procedures for general
recommendations. If the pond is harvested too late, the percentage
of oversize fry and cannibalism are greatly increased. If the pond is
drained too early, an unacceptably low percentage of fry will have
developed to the free-swimming stage.
(2) Failure to adequately disinfect the pond or prevent entry of
wild fish. Abundant fry in the pond at the time brood fish are
stocked may result in a 90% reduction in the harvest of target size
fry.
(3) Draining mishaps, such as broken filter screen, prolonged
drainage time as a result of clogged filter screens, accidental com-
plete drainage of harvest sump before fry collection, etc.
(4) Diseases, parasites, predators, and poor water quality are in-
frequent causes of depressed harvests under the management
scheme suggested.
-50-80 cm -
,- 90 cr
I I .....=. ............... .................................. ...... ................. .......... .......... i l I
If 6111'' 0#' 11'' #'' 1111111''' f' ''v [ ''oilo'1v1 ''f v'l'''''l''1'11 '11 ''M''of' '111''''' 1 v'1 '''11 '111111 It '''1111 ''.'''' ' i'li ' I
910 c
40-50
i-
Support bars
END VIEWSIDE VIEW
.--- Handles -..
1- PVC floats--
O
Marine plywood
end plates
t a 3-cm mesh -
metalic screen
FIG. 4. Floating grader used to separate fry.
[6]
1%1
II
I
I-
a-
-~ ~
U ~
-4-
A
.~ ~j~4~X 
~
FIG. 5. To prevent cannibalism on recently hatched fry, all fish must
be eliminated from reproduction ponds by thorough drying or disin-
fection (A). A large mesh netting is draped in an earthen or concrete
harvest basin (B), and the pond is then filled with water and brood
fish are stocked. Then 2-3 weeks later the pond is slowly drained to
the harvest basin and the netting is lifted to remove the brood fish
(C). Fry are skimmed from the water surface with a scoop net (0). If
fry too large for sex reversal are present, the harvest is placed in a 3-
mm mesh grader that is floated in a fine mesh hapa to receive the tar-
get size fry small enough to swim through the mesh of the grader (E).
It 50 kg of female brood fish are stocked in a 500-1,000 M
2 
pond, pro-
duction of target size fry is approximately 50,000 per cycle.
PREPARATION AND ADMINISTRATION
OF HORMONE TREATED FEED
Concept
N idhials, tiex arc ti alixlei red ti hiaptx xxlel tllex wxill li glxtii :1 hor-
lixtc. 'I lie hoi llic ti-iatililiit cailili ll(ie xisipcimh itt.lil tihe
%doit ili~t filtItilictuiuial uiiiic fiih T[lt precc l i uiiii iiit xx 1li the
ptaimexicl bh 24 to 2l
3
(dix 11(1 mliotilitek 3 ldl-to I xxitx at il
1) 1 t i ii cIto'2 t o :t I i i h himl i i ilk o ciir fc : to I i\\lt til lilli lit Ii wi) di
N I il! I I i l c l ldatl I 11.11 totial le l- uth xli. tl Niiii iix Iht.
Ix ) iiii h I i xx' I te 1l l hnt t a tlil t x l d cI ii t i o l icI 1-\/ etol 1 x I I i'dIiil
I lii t\ lt i l 1( 1 t a t Ilia 1(), t o 11 11)1I 
iIi j iittI J pl i liIi t t hI h olt
f 51) will( cl xx 1 i ti llllili c t atel depth lol e l( traim.ti
anti csotcole thellIol(1t ( em () axhltitat l t) 1c ixiii re rtx i xant
is ini llltitil lixte biiiuit lilt
1
k disx e a iiliel litO h to 90/ xti eall
K x " th e~l oillonxe inxto. h 'c ,alaclo-o-loi oltoli
(4) A1 ic iiittxic tIcd i 2 to 5 g ilicrel is iitcti idIIkg alidito -ap
1) Fal i pond thaeitx ie 1 to 2th nio'pn alirxa per hbitth ii'
-ir1k
50,000 fry, including feed preparation, fry handling, feeding, and
harvest.
Procedure
The sex reversal operation is divided into seven major activities:
(1) PREPARE THE HORMONE IMPREGNATED FEED
Selection of nutritional ingredients
The hormone-treated feed should be of good nutritional quality
and highly palatable to ensure ingestion of the required amount of
hormone. Crude protein levels of 25 to 45%, at least half of which is
of animal origin, and vitamin and mineral supplements are recom-
mended, but 20% protein feeds have been used successfully A com-
mercial starter ration for fish is ideal, but a finely ground, balanced
shrimp or fish feed is acceptable.
A base diet also can be prepared by mixing a finely ground com-
mercial chicken or swine ration starter with fish meal. The chicken
or swine starters generally contain mineral and vitamin premixes of
adequate quality, and the fish meal increases the protein content and
the palatability of the diet.
The dry ingredients should be sieved to remove particles that are
too large to be ingested by the fry. A 0.6-mm mesh screen may be
most appropriate, but no obvious problems were observed with
sieves made of commercial window screen.
Alcohol-hormone solution
A sufficient quantity of the alcohol-hormone solution must be
added to the dry ingredients to ensure an even distribution of the
hormone. Normally, about 0.5 liter of solution is mixed in 1 kg of
diet. However, the small amount of hormone needed (60 mg/kg of
diet) can be dissolved in a much smaller volume of alcohol. It is
therefore practical to prepare a concentrated "stock" solution of the
hormone that will subsequently be diluted to a greater volume be-
fore mixing with the dry ingredients. The alcohol in the stock solu-
tion should be 90-95% ethyl alcohol, but need not be reagent or USP
grade; glycerin may be added at 0.5% by volume to render the al-
cohol unfit for human consumption. The alcohol into which the stock
solution is mixed may be 80-95% ethyl or isopropyl alcohol.
Stock solution of hormone: Dissolve exactly 6.00 g of methyl-
testosterone in exactly 1.00 liter of 90-95% ethyl alcohol. (This
quantity is sufficient to treat approximately 300,000 fry)
The stock solution should be stored in a dark bottle. It can be kept
at room temperature but preferably under refrigeration. Shelf life is
at least 3 months.
List of ingredients per kg of diet
Alcohol-hormone stock solution exactly 10 ml
Ethyl or isopropyl alcohol about 500 ml
Dry ingredients 1,000 g
Procedure for mixing ingredients
(a) Prepare the ground and sieved dry ingredients.
(b) Mix the hormone-alcohol "stock" solution with the alcohol.
(c) Add the above solution slowly and mix with the dry ingredi-
ents.
(d) Allow the alcohol to evaporate in an oven at 60'C or at room
temperature with no direct sunlight by spreading out the mixture to
a maximum thickness of 3 to 5 cm. Mix lightly by hand 2 or 3 times.
(e) If the mixture has been completely dried in an oven, it should
be sealed in airtight containers and/or stored in a refrigerator to re-
tard bacterial or fungal contamination. If dried at room tempera-
ture, the shelf life of the diet may be extended by sealing in plastic
bags when it feels dry to the touch, but before all the odor of alcohol
has disappeared-usually 6 to 12 hours. Refrigerate or freeze for a
shelf life of at least 2 months. The diet can be kept at least 1 week at
room temperature.
Quantity of treated feed needed
Sex reversal usually requires 250 to 400 g of treated feed per
1,000 fry. All the feed needed for the entire treatment is usually pre-
pared before treatment begins.
(2) PREPARE THE POND FOR SEX-REVERSAL HAPAS
Verify pond and water characteristics
Chemical composition of water supply should be similar to that
indicated for the fry production pond.
Cooler water temperatures (20 or 23oC) reduce fry growth but do
not negatively affect sex reversal.
A harvest sump is not required, but the pond should be drained
at least once per year to eliminate escaped and wild fish. (Free-
swimming fish weaken or even make holes in the net enclosures
while attempting to get at the feed or the fry)
Fill pond if necessary
The convenience of chemical fertilization or manuring is site de-
pendent. (Increased abundance of natural food will increase fish
growth, thus increase feed requirements. However, manuring may
be needed to reduce inorganic turbidity, and fertilizer induced
plankton blooms may improve the dissolved oxygen dynamics.)
Chemical treatment for control of aquatic invertebrates, as com-
monly used in primary nursery ponds, is usually not necessary.
(3) SET UP HAPAS
Number and size (See "Input requirements")
Location
Easy access to hapas is important because feed must be given at
least twice daily The hapas should be at least 30 cm above the pond
bottom to permit water circulation and prevent turbidity For these
reasons, net enclosures are often tied to wooden walkways or piers
that extend a few meters from the shore.
FIG. 6. Where pond water levels fluctuate, hapas can be secured to a floating frame.
[8]
SIDE VIEWEND VIEW
--- ---- -- 1 V
Securing the enclosures
The hapas should extend at least 20 cm above the water surface to
prevent the fry from escaping. If the water level in the pond is con-
stant, the hapas can be tied to stakes driven into the pond bottom.
However, if water level is likely to fluctuate, it may be advisable to
secure the hapas to floating frames, thus ensuring a constant dis-
tance from water level to top of netting, figure 6.
Covering the enclosures
The hapas generally need not be covered, but in some regions
protection from predatory birds may be appropriate. If a covering is
required, it should not completely shade the sunlight, which may be
an important factor in disease prevention.
Accessory equipment
Floating feeding rings made of 3-cm-diameter flexible tubing
should be placed in the net enclosures to reduce the loss of treated
feed, especially on windy days. A circular ring with a diameter of 50
to 80 cm is adequate for 10,000 to 20,000 fry in a 3- to 5-m2 hapa.
Submerged feeding platforms are sometimes placed about 40 to
50 cm below the feeding rings to further retard feed loss, but their
importance is incompletely understood. A plastic sheet should be
preferred over a rigid platform which causes excessive wear of the
netting. Fish have been successfully produced in hapas with no sub-
merged feeding platform.
(4) STOCK THE HAPAS WITH GRADED FRY
Stocking rate
Suggested stocking density in net enclosures is 3,000 to 5,000
graded fry per m
2 
of net enclosure. Fry have been successfully sex
reversed at lower densities, but the enclosures are inefficiently uti-
lized and the relative abundance of natural food may endanger the
treatment if the diet is not highly palatable. Higher densities may
also be possible, but competition for feeding space may result in
more undersize fish that have not consumed sufficient hormone.
Counting
Before treatment, fry are too small and delicate to enumerate
with methods based on estimates of average weight. The following
procedure, based on a visual comparison against a known standard,
is recommended:
(a) Select two identically-shaped and light-colored containers with
vertical sides and flat bottoms (about 30 cm diameter). White 20-
liter plastic buckets are convenient.
(b) Add about 5 cm of clear water to each bucket.
(c) Count 1,000 fry into one bucket. (Fry segregate by size so they
should be momentarily crowded when selecting a representative
sample for counting.)
(d) Place the second bucket next to the "standard." With a shallow,
fine mesh scoop net (about 10 or 15 cm diameter), transfer fry to the
second bucket until the fry densities in both buckets visually appear
to be identical.
(e) Stock the fry in the second bucket into the treatment hapa, and
repeat the comparison process.
This method of comparing against a standard is rapid and rela-
tively accurate. In one hour, two people usually process 50,000 to
100,000 graded fry. Error is seldom more than 15%.
(5) FEED THE HORMONE-TREATED RATION
Feeding frequency
Two to 4 times daily during daylight hours, 7 days per week. (No
detrimental effect was noted when 1 day per week the entire daily
ration was given in a single meal.)
The daily ration need not be divided into meals of exactly equal
weight. Normal procedure is to measure the daily ration for each net
enclosure and to visually estimate an appropriate fraction of that to-
tal for each meal. The daily ration is most accurately measured by
weighing, but a calibrated measuring cup is adequate and less time
consuming.
Feeding rate
15 to 20% of fish weight daily until fry reach an average length of
15 mm with gradual reduction down to 10% of fish weight daily until
the end of treatment.
Calculation of the daily ration
The daily ration per hapa is based on the desired feed rate, the
known total number of fry in the hapa and the average weight per
fish as calculated from the known or estimated length. The approx-
imate length-weight relationship of fry is:
W = 0.02xL
3
Where:
W = weight per 1,000 fry in grams, and
L = average total length of fry in mm.
Based on the above relationships and recommendations, the sug-
gested daily rations, according to fish length, can be summarized:
DAILY FEED RATIONS ACCORDING TO FISH TOTAL LENGTH
Average Daily ration Average Daily ration
length per 1,000 fry length per 1,000 fry
8mm 2g 17mm 13g
9 3 18 15
10 4 19 16
11 5 20 17
12 6 21 19
13 7 22 21
14 8 23 24
15 10 24 27
16 11 >24 30
Fry may double their weight in one week. Rations should there-
fore be adjusted on a daily basis. The ration for the first day of treat-
ment is taken from the above table after measuring a subsample of
fry at stocking. Subsequent daily increments in length are estimated
from known or assumed growth rates. Common growth rates at
temperatures of 25 to 28'C for the stocking densities and feed qual-
ity previously described are:
Length of fry
8 to 12 mm
12 to 17
17 to 25
Growth rate
0.2 to 0.3 mm/day
0.3 to 0.6
0.6 to 1.2
These figures should be used only as an initial estimate of ex-
pected growth because natural productivity, temperature and feed
quality are highly variable. At each specific location, growth rates
during treatment should be determined by weekly sampling of fry
during the first few cycles. Once growth rate under a given set of
conditions is known, a daily feeding chart can be developed to facil-
itate long term management.
EXAMPLE OF 2-WEEK FEED CHART
Assumptions: initial fry length, 9 mm; growth rate, 0.3 mm per
day up to 12 mm length and 0.5 mm per day thereafter. Method of
calculation: first determine the expected length of fry for each day
according to the above assumptions, then round the length to the
nearest mm and find the daily ration from the previous table "Daily
Feed Rations According to Fish Total Length."
[9]
I )a.\ F ih Iii 4th I )ilxI
I itititi
1111 g1,00)1
1 9.0) 3
2 9.3 .3
.3 9 6i 1
1 9.9 41
5 10.2 1
It It).5
7 10. S
I )ix FI Iii l ti4l I ikil
I itiiiii
111111 gl 1, 000
8 11.1 5
9 11.4 5
It) II.7 6
11 12.0 6
12 12.5 6i
1:3 1.0 7
14 115 S
(6) HARVEST THE SEX-REX EBSED FING.ERLINGS
'Ilreatmciit dutration
Frx cni hi ifleeIt ix (lx xex rex is xiii li 2(1 dax liut ocioin nxlk il
oiilx\ 95/( of tlie fr, ihxelopj axs 1lictiiitx pw iiilixes Sex i exetl ,alxie
Cexxix inoii iiiixixstiiit xx ien the ti eatmitt iiiatioii ix 2.5to 28 iax,
NIlinirni aicceptale size
Aftera 2.5-to 28 dax treatimeit fexx ffx tie lexx titalIt 111i1. Floxx-
ix ii if iioi i tima 5~/(iare 1.3nm oii exx tliiix iliix M(iliaxiiiislild li
eiulled hecauii a vi ixs 25(/( oi them ia he teitalix. 'I'l( he' 41 di
that haldbieil iixed priol\ xiiiii Ii inox v ei xi/ie fix for xex ,rext ii
'At the endi of tieatiliit the axcilg e xxeighlt of' the Ii iatid fixh is
iixiallx\ 0.1 tii 0.3 g4. \t tis xi/e it ix illoi i iiiiixiiiit aiiil aceiiriti
tio ixtinati thei fitlial nniiliitr iif, xti? xi\ il,, fisht Ix (lix iiiii tihe totakl
fiinal xx eji it ofi fixli ill the hilpa hx iti ax il wi ii it oiii a xiiliiili.
A 1-ip eiittiix xiiliaiple (ahiiiit 1001 to 150) 14) caiili iilibtaiiiii
iiilx after ti tlxt er\iiiiii all the fixsh intiiaciii iei of lii net ii dii
xiiii. Fixst iif tis xizi ii' xtill deilicate. flux xhiiuldl lie xxi 4hd if)
tiricl buets xxith xx atii toi iediiei the tutu i oit of xx dii
mii tiii il l i Ill()]( ( miitt ii l cm ix - i it ( it i ( I 1 ix tliii k, of li i
of, the i(itllxd \Nii il ) el cifk \N liciee i iiiiaieiptil ighl per-
iiiitage~ ot iiCilae l iii 1111sr pon wasx( () 111x ix ii to c(iti iiiht i o ii
Expected Results (Fig. 7)
I Iormil(ne Treatment IPhaxc
Axeii . i xi/i afti i 25- to 28 d(ix\ tieitiiietit
\\ei-lit: 0.1 tot 1.3 (1,
I ai4tii IS to 25 mimt xx tl at leaxt 95'/ 4reatci (hii 14 ii1iiii
SeconciarN nurser\ ph~ase
.Silli il l (l1in 111cr iti i 60 to 90
\iragai xti after 0 toi 1( it. x: I5 to 25
A!
I \1v, xi i ~ lm lv Fi ~ x iN. IN , Vx vxi xii
Iiikei x\fill xx tc puti ,i
Itiitdiiei of i xti fi lixkc
4250) )
1:31 4
550),,
180)
-1,250,,,
(7) SECONDA~RY NURSERYI OF SEX-REVERSID FRY
Aftei xex rei ixeal, tlhe lixii iiiax hc xtiiekiii iirietx liii groixxiiit to
iiia ket xi/i, hut xiirx ix al maxlii be 4l xaiialile. Fir 4iiatet prc
(i(iiailitx andi more e-fliceit uitilization oiif faiilitie a ii wiiid if xie-
iiiiiiarx niirxii x ill piindis is xtil 4itiei. All xx ilil fixh shuilid laxi lieen
iliui teoiiii the piind, andi produtivei plankton eiiiiitx
xhliid hi ixtaliliii prior to xtockin 4 I piiiolx I\ itif l p )iii piiix,
xiii -im xix1-r ai iiii iti sirixu ea loiix a.5 tii 1011. lIt piiipeilx ia
ail inn xei xl pimdx xliid i iixiallx x ariix ftiiinl (it tii 901/r ox ll i
axverage of* aboiut 75/(
Alt iflietix cii iie x tiechiqiue ix to fill atuil fii tilizi a (lix\ or (ix"
iiifeetiei piid xx itli tilteiv iiixxatii 7 tii t) (lax\ x piiir tii xtiiikin 4. Afteir
iiiitial fei tilizatiiin xxitli eltiekei litter at at iate of' 1,500( tii 2,0)1)0 kg/
haiii ofi ii iiiaiiiiie ft oiiii cattle oii-xxxime it a rate of:3 (100 tii 5 1000(
kg/hai the tiiirxerx poitdis ,eiuiialkx dexiliip planiktii biliiiims xxfili
Siuelti isk x ixiiiitx of 30) toi 40) ii in .5 tii 7 (lax x. iria 11 1.5 k,,,
lia) caii xiippicitteittheii initial iiiaiinie applicatiii. [5\ c all li
xtoekecd at a t ii of, (110) to 125,000( per iietaie. Itii it nuii xii
tlii pildouiiilid be ttalutied xiiklx att 1,(0001 k 4 hai. A xiippliuittl
ieCd at .5% iif'the fixsh biiomaxx dailx iinpime ox x i i tit huit ix nit el it-
ll f(ii high) xii xi l.
if thei i asigniificanit pfiililitx that liiitiviatid tilalpia iliax eiitei
thc iiiitxsei piid. at xliiilpl iif' abiiout 2(1(0 tivi eti fixli xhioulid li
B!
> Ai-
FIG. 7 Pictorial summary of sex reversal procedure. 
Tilapia fry can
be sex-reversed in fine mesh hapas (A) that 
are tied to fixed stakes
(B) or to floating frames (C). Fry less 
than 15 mm in length are
counted into the hapas by visially comparing 
against a standard of
known number of fish (0). They are given 
a hormone-treated feed 3
or 4 times daily for 4 weeks (E). After daily 
feeding for 4 weeks, 97-
100%o of the fish are phenotypic males.
COMMON QUESTIONS 
ON THE PROCEDURE
Tiii frjl s desciX fixriptiion XX\-its iii aiii't tii faeifjiate 
fiel
1 
aipiiii
tifiealtiiii amid at'tefpta lie uiot~ilfi'tiiiuS oi- itel jliatix ex tio Xpetifie at'_
tiic feel Xloiiiifii cd ii Thoris secini if )incl' Xude 
Xt\oIaesist xaiilliaXi tt
Xpfit'' 10if It ini(iatiiiiX inti il i)Xi'iaii filill iiiaiiagt'liit.
Hlow is the Effectiveness of tile Sex-r-eversal 
Treatment IDeter-
minled?
illatlioni iii til !i'iitii papiiia iii' ibx sai iht'uou the 
fishi tii examine tiie
e'iiXI' femoilieX flax t txX o illtiepeulitlit orifiees il tiit 
gellital pafpiffa
Ito'I retlease~t of' iggx an oi liilit' XX ii ii'i ill iialeX 
tihe dti c ar rx ii
Xpterml aint Iii oill iiit('i liaiXl~ iitl texit hu iiigiI' 
ii iigt'ita) oi
fice figum 8. There t're t two ( ililtipil diil(i 
litagcX XX itf ti ttetil
iiiii(: f1 Bllx w' Xitii experi'iiee Xvisul texainuatioisiI 
not tciiis-
tenitiy iri'iaible liitii tlit fishl fluXt 
groiwn i tif a XXeigiit ii aiiiiit 20) g,
itild i2)ltrings tllrl ij, iiathiisiiiatiiiio h 
gouiad.tcolitiflpaitlla
tel-i k iai i ~ jiptli to Il' Ii l itlilb t lX n i i'll 
if)I teli iai ik a it'i femilt'
\\ itIlI o\ i cs I X c rII (,~I ie 's"' 10 1 )h tit'a 
IIl I)ii'foX(' tIf is' tecIf(I I iqIie
illilx be' sati sfaetorx\ e pt'&i ai of iiri i i X 
i tre i(ate d fi 'lh ire
rile ii''i ll 11111 Xt'rX poiiih to aitfx dic''(dO imeiii 
or e'aly iit Xtalgt'
ibefore s'tocking ill at iloXot pond1.
Whenu sex-rex eis Xiv a rxwe stocedt in to 
gr'(mX 1 t ponds(1o01 sold at
ii eaixd Loge it is oftel 11 important to ex a~ii 
te th e sex i exersiaf trilt-
iIleit ealrliei anil ore i'i'iiai)x Ill thalt 
ealse exainationi of the go-
niad of at suitil aipie of 0 eaitet f is isI 
the suiggesXted( ilttho( . The
gI iinit is cx(\alliied ill 1(1osceiiial l 
aind ooc\ ti' aie r'iead ily obs~erv ed
jif th miadi is aii OXaiv T.he ('latiX tix 
liir gi ioinifild e ctes ieeoiiie
C('X ii10i ilolI f)\ ioii if the gooai is Xtaiiieti 
i)efoi1- et\iiliitioll, figure-
9.
A St \i \t~~ kmi Di iliiliiN )\ offi iIliit 
Ii
ait Aftei Xex \ r ' alI reall at sitiiipie 
iof' fish ill a iti trlled eni
Xirouiniit to at siiit ize oi/'14 to 5 tini. 
(InitiaX irtar thein to 8
to 1 it) tii itil tihet crew Li ilX tXp)Iiclice it
ib) Pre',erxe thefit' fi it 7 to 101c forioifii 
soluition ii0forat le'ast 2
\N) 'ekX, (inThe frinali tioiugihetos thi' tissXIue al 
id m akes removin)Xal of'the
golif i ilitfil eaxsil!.)
(ct ( )p 'ii the is h, extract tile thl l gol 
iai XX iih lie if' li iiig tihe (liii
Xf] (tioi) XSide iii tlit abd ili i i i t i andil~ 
plate it il i a giXs Xslide.
lo Pt at drof) o)'i'toifi l 00' staini 
of) thie golladi iiif fighitiN
Xqifiiahi it XXit it f as iaXX to ip. J ii 
Xtaiii eall fbt fpurcthased ili
p~rep~ared' foi o or maide b\x adinug 0.5 g of'in-lube 
to i00 )ill of 45%
FIG. 8. The phenotypic sex of tilapia can be determined 
by visual ex-
amination of the genital papilla once they have reached 
a minimum
weight of 15-20 g. The urinary duct and oviduct 
exit independently
from the genital papilla of a female (left), while 
in males (right), the
ducts carrying seminal fluid and urine unite 
internally and exit via a
single urogenital pore.
(c\ fterit acouple4 of' li1ilIlittx. t~all(C iit 444c (_Iold alt 50X llia44lifi-
cttloil IT i(' 'ixil I i i aprt I I I I p t k c iliC If it1i iXIX packe I oocX tx~ are4
no(t fo111)d. ( )ccax'iollaiX, all o(XI ('tet'X wXill lic 01474(1 d \\( XiII stilt-
(('1(41 oo47XtesX 1C%%, if .1lI of til(' i xi xii lo' iw )IpodcitI47t ti 1(11
tI ioil fecotale'x
'Filt' rl~iilit\ of, tilt( 47tilliatt' of' percen7t'(I llc ill4 i batch171 of'
tre7ated( fishi i filictiollof, the' lllillthci of, fishillII the saiIIIIeit' 111C7
9.51/ 7oinflidecin4 It47rX ti 9.)t4 asXXilimct hat11, tilte tuc ax t'1a4( is
XXithlo tlic( iaio5 indticatedi) i)(colict' tighlteri s tilt' samlsi Xill-
cre(ases 'X 1 01 2(0 to 1,000( fish. II ax tragc pecen' t matles ill iample (114
is 95%, tihe ('(1 Ct (If'sml size o()4 i l/' tiit'(( ofitec 11('i litti~a X kII
96
97
98
99
10(0
9,5(/( conifidenilce i(tclX iI
N =100(
90.1-98.9
91.5-99.4
9:3.) 099. 8
94.6-99.9
96.4-1001
N -200)
92 :3-W9.3
9:3.6-98. 9
95.0-99.4
96.4-99.9
9S.2-10(0
N .51)0
94.11-975
95.2-98.2
96.4-98. 9
97 7-99. 7
99.3-100(
What Pei-centage of Fernales can be Tolerated in Crowout
Ponds?
Se\ revcrsid, its described it) this publication, is gcllcrillk llot
100(/, cill-'Okc. \lost 1,1-\ ill"est illore holillolic t1litli I-c(Illin-d to
ildii( w sc\ l-1,\(,ysitL but tcedill(-, 1whiMor, collipctitioll, dis-
cilsc. ("clicticEactorsilla.\ Ill('-
\cllt
treated f'Ced. 'I'lic pci-C-clitit'gc of plicliotvpic I'CillitIcs after ti-ciltillent
is oftcli less thall V4, but oil occasion niii\ bc its high its 5(/. Some
oftli(isc lCillides illil\ be 'Stcrilc. blit others \\ill )tit\(, llorillill rcpro-
(Illokefillictioll. Ill \\itrljjcrC.lillIiIt(.,,,
1writtm-esot'itt lcitst :25'C(7TF), the\ min spimil ili:3111olitits. Lim-
ited reprodlictioll late ill the gro\\ollt c\c1c hits little lic"'ittkc clicct
oil "l-m\t1l, blit tit(, dallger ofstlllltillt incl-citses prilliarik its it 1,11TIC-
tioll oftlic 1111111ber off'CillilIcs itild tll(i durittioli of flic growout c\cle.
licstilt's ilre \ ill iiih1c, dcpclIdill'_' oil illitiliu'relliclit practices and CH-
\ irolliliciltill coilditioll's, but tit(, 101lo\\fil(_1 hillits are 1 'i\cll it's it "'un-
('fill "I Ilic-of-thillith ", for \\iIrl1wI- clilliatcs:
sampilet size (N)
21)
,50
100(
2(1(
.5(0(
1 00011
X. ill(, purCilt 1(1.1144 illit . Xlilfpic is:
'r 4
jP4
'7 ' .'~.
'4
47 ' -
.4 *'' "T ''~ I. A'
~w
I
S. * LVr
7
* w
1'~
I4~.' ~
~'
4
',j
~" I
FIG. 9. Qocytes at various stages of development are apparent in the
ovaries of 4- to 5-cm females (above). In phenotypic males, oocytes
are absent from the testes, which appear granular in texture (below).
Tillic ft-oill wX rc\crsill
to liill*\(,St oflood fish
:3 moiiths
5 motiths
7 illolitlis
10 Illolillis
\laxi lilil "(il III (tIl11114
(If plicit( i Jlllit'Xs
7- 10f(
:3 -5 (/
I -2 %/
01/1
Pol\ cliltill-c \ ith all ellecti\ c Ill c(liltor llliti ,'iltcs tit(- c ill]-
pact ofit fe\N spimilim" felilitIcs hilt illil\ lilldlik complicatc the lo-
of poild Illitila"Iciliclit.
What Hormones (],tit lie Used?
S.\ litlictic compounds an, dc\clopcd its allil-
holi( plilliallik lot TilliscIc dc\clopillclit! alld ol
il("clit's. Some 11111"t 1willieck-d ilitrillillisculalk Mideothers lllit\ be
takcii oial1v \1(-tli\ Itestostei oile iuld 011\ inItestostcrolic cilli bc
takcil orillk alld arc abolit (i(Ilialk potelit its all allilbolic aild its it
illitscillillizi -lig, il"'clit. liccitlise of tit(, ("clicrill a\iiilabilit\ itild cfiCC_
tkciless of, 111011\1WOosterolic. litt1c research llas heell colldil(ted
()It SC\ R1\Cl111tl of tililpia \Nitli 1clatcd horillolics. IIo\\c\cl., prclimi-
iiitr\ \ ork bits (Icnionstratcd flicir fCasibilit\. Thc most lilcl\ cim-
didiltes to]- imcsti"'ittioll are k-stostcrolic collipoillids that call bc
taken oralk , arc strotil(, initscl If ill izing it"'Clits, ;kill[ ill-(. olIl\
,ollible ill \\ -1*.
-iit(
Ai-e tbei-e less c\pensive altei-natiNes to alcohol a\ailable?
Thc 500 
ill] of ctb\ 
I alcoliol 
1)(.1. 1,111 
ofdict, 
as ,11( ... 
ested ill 
this blo 
-chill-c' illil\ cost its lillich it,, tlic hol-Illone itsch'. lllortllllatck littlc
rcscarcli bas 1well perforined oil Silitilble slibstitlitcs. Sollic l-clc\itllt
pomts ilrc:
I ) It is probilldc blit lillpi-mco that lesser quillitities of' ('tit\ I A-
cohol \\otild not 1 the efiCctkelless of tll(i ti-catillelit.
(2) 111 mam comitrics, potaldc (-tll\l illcollol is llllil\iIilablc ol. is
illadc illore expellskc its it reSolt of Special taws. Kth , \1 illcollol dc-
liatilled with "d.\(,(,]ilw 0.5% call be Side].\ lised it) thc pi-cpa-
ration ofil Sex-l-c\c]-sill diet.
(3) Isoprop\ I alcohol, its inentiolicd ill the t('\t, is it 1)1-o\ (ill Sobsti-
tlitc lot 011\1 illcollol.
1) Meth\ Itcstosterolle is soluble ill solne fish oils. An addit iollill
99.9%c
C99. % 4
98.4 cl
97.6 cl
96. 7(7
96.3 (1
95% confidence iiit(,i-\iil
benefit from spraying a hormone-oil mixture on the feed could be
increased palatability of the diet. The increased risks, however, are
the danger of rancidity from improperly stored feeds and the possi-
ble loss of hormone if the oil floats off the feed in the water.
Can other methods be used for fry production and sex reversal?
As previously described, large numbers of uniform-age fry can be
produced by draining a reproduction pond down to a harvest basin
about 3 weeks after stocking brood fish. However, other techniques
may be more appropriate under certain conditions. Suppose water
is scarce, or a good harvest basin is not available, or a large number
of fry are not needed on a single harvest day. In those cases, the re-
production pond can be partially harvested on several occasions, us-
ing a fine mesh seine with the lead line held just off the pond bottom.
Partial harvesting should begin no later than 3 weeks after stock-
ing brood fish. Depending on the frequency and intensity of the par-
tial harvesting, the pond may remain productive for 2 to 3 months.
Eventually, fry production is reduced by the cannibalistic juveniles
that escape capture.
The principal disadvantages of partial harvesting are: (1) it is more
labor intensive than the complete harvest technique as a result of
additional sorting of fry by size and removal of unwanted by-catch;
(2) the increased handling stress produces greater harvest mortality;
and (3) as the biomass increases in the pond, slightly reduced
growth will increase the average age of fry and the risk that the go-
nads of more genetic females have already differentiated into ova-
ries. (The maximum acceptable length of "target size" fry obtained
by extended partial harvesting has not been determined.)
An alternate method of fry production is: Brood fish are stocked
at a density of 3 to 7 per m
2 
of hapa with a sex ratio of 1 to 5 females
per male; fry (and possibly fertilized eggs) are collected every 5 to
10 days. With the more frequent harvest schedule, the percentage
of free-swimming fry is greatly reduced, and would therefore be
practiced only if incubation facilities are available. If hatchery
equipment is available, the brood density is set at the lower end of
the range and the proportion of males is increased to improve the
probability that most eggs will be fertilized. A temporary resting or
recovery period for brood fish between cycles apparently improves
fry production in hapas. Promising large scale production tech-
niques are currently under development at the Asian Institute of
Technology in Bangkok, Thailand.
Fry production in hapas has several advantages and disadvantages
over the open pond technique:
(1) Undrainable ponds can be used.
(2) Water is conserved because fry production ponds are not
drained as frequently.
(3) The pond may be used simultaneously for other purposes, such
as nursery or growout.
(4) Brood fish are used efficiently. (Harvest mortality of fry is
lower, and fertilized eggs taken from the females can be incubated
in hatcheries.)
(5) A relatively deep harvest basin is not required.
(6) Fry production in hapas is more labor-intensive because fry
must be harvested almost weekly, instead of every 3 to 4 weeks as
in ponds.
(7) Spawning hapas and incubation facilities are relatively expen-
sive. The use of spawning hapas more than triples the total area of
hapas required for the fry production / sex reversal operation.
(8) Brood fish are more susceptible to predation and theft.
What other containers can be used to sex reverse fry?
Tilapia fry can be sex reversed effectively in any container that
maintains environmental conditions appropriate for growth.
Aquaria, stainless steel troughs, plastic pools, and concrete tanks up
to 30 m
3
have been used, the latter on a commercial scale in Israel.
The problems are:
(1) The amount of hormone-treated feed administered requires
that the containers be cleaned daily and that water quality be main-
tained by frequent water exchange and/or aeration.
(2) Disease and parasites are more problematic under such con-
ditions, especially when the container is shaded. Mortalities of 50 to
80% are not unusual.
(3) Labor and start-up costs are generally higher.
Is sex reversal affected by the frequency that hormone-treated
feed is offered?
Feeding frequency during sex reversal is a subject of great prac-
tical importance currently under investigation. In yet unpublished
studies, tilapia were effectively sex reversed when the hormone-
treated feed was offered only 6 days per week and when the daily
ration was divided into only two meals. However, further studies are
suggested before these practices are recommended for commercial
operations.
Can Sex-Reversed Tilapia Be Used as Brood Fish When They
Mature?
A few genetic females may consume so little hormone during sex
reversal treatment that they develop into normal, functional females.
Another small fraction of genetic females could be sterile individuals
with ovo-testes as a result of insufficient amounts of hormone. Mega-
doses of hormone also could cause sterility in a few fish, but the ma-
jority of the sex-reversed fry develop into reproductively functional
males. About half of these fish were genetic males from the moment
of fertilization, and their offspring would have a normal sex ratio of
1 female:1 male. The remaining sex-reversed fish can function re-
productively as males, but they remain genetic females, and when
crossed with normal females, all or most (depending on the species)
of their offspring would be normal females. This is the opposite of
what is wanted for monosex growout of tilapia. Therefore, although
the majority of the treated fish are not sterile, a producer would take
special care to eliminate all sex-reversed tilapia as potential brood
fish.
Do Sex-Reversed Females Grow More Slowly Than Genetic
Males?
When sex-reversed fish reach market size, there is no obvious bi-
modal distribution (two size groups) as is the case in mixed-sex cul-
ture. Many factors related to the phenotypic and genotypic sex of the
fish determine its growth rate. The relative importance of each fac-
tor is incompletely understood, but the net effect is that, for all prac-
tical purposes, genetic females that have been sex reversed to phe-
notypic males grow the same as normal males.
We noted that when sex-reversed tilapia are harvested after grow-
out, a low percentage of the fish (less than 1%) have distended ab-
domens. The urogenital papilla are phenotypically male, but the go-
nads may be filled with a watery liquid containing few or many eggs.
These rare individuals are probably sterile "intersex" fish with ovo-
testes. The liquid accumulation may be related to the lack of a con-
tinuous oviduct leading to the exterior. The fish appear to grow nor-
mally and have no other sign of abnormality.
How much does the hormone cost increase fingerling cost ?
Seed production requires many inputs, with the cost of individual
inputs varying from country to country. It is, therefore, impossible
to give a simple answer to this question, but, in general, the cost of
the hormone is a minor fraction of the total cost of producing mono-
sex seed of tilapia. In the United States, the cost of the hormone in
1989 was $26 per 10 g, which is sufficient to produce nearly a half
million sex-reversed seed. In many developing countries, it may be
more difficult to obtain the veterinary grade hormone, and the grade
approved for human medicine may cost three times that value.
In Ecuador and Honduras, the total cost for fry production and sex
reversal was equivalent to $5-8 (U.S.) per 1,000 fry in 1987. The cost
of the hormone treated feed was about 10% of that total, and the cost
of the hormone itself was about 10 to 20% of the cost of the feed (that
is, 1 to 2% of the total cost of producing sex-reversed fry). In most
[13]
countries, the approximate range of costs for the major ingredients
needed to sex reverse 1,000 fry is:
Ingredients
Basic feed
Alcohol
Hormone
U.S. dollar
0.30 - 0.80
0.20- 0.30
0.05 - 0.25
Hormone costs would be reduced if a lesser quantity were re-
quired. Several investigators have shown on experimental scale in
clear water that tilapia can be successfully sex reversed when fed at
only 10% of body weight daily with a ration containing only 20 or 30
mg of methyltestosterone per kg of feed-that is, only about one-
fifth as much hormone as recommended in this publication. How-
ever, we would not recommend these lower levels until they have
been field tested on commercial scale in water containing abundant
plankton. This precaution appears even more appropriate 
when con-
sidering that the cost of the hormone at suggested levels is only 1 or
2% of the total cost for sex reversal.
What Effect Do Hormones Have on Humans Who Eat Hor-
mone-Treated Fish?
The justification for the use of methyltestosterone to sex reverse
fish intended for human consumption is based on considerations of
the total quantity of hormone consumed by the fish and its rate of
elimination from the fish after the sex-reversal treatment is sus-
pended. The total quantity of hormone consumed by the fry during
sex reversal is small in comparison with normal therapeutic doses for
humans. The minimum recommended daily dose of testosterone for
androgen-deficient human males is more than 100 times greater than
the total amount consumed by a tilapia fry during sex reversal. In
reality, most of that small amount is metabolized and eliminated be-
fore the fish grows to a marketable size as the liver converts it to
more water soluble compounds which are excreted in the bile and
urine. When methyltestosterone is orally administered to fry during
sex reversal, 90% of the hormone is excreted within 24 hours, and
just 3 weeks after the hormone diet is withdrawn less than 1% of the
hormone remains in the fish. During growth to a marketable size,
the juvenile and adult fish continue to excrete the remaining hor-
mone. At the time of harvest, the quantity of dietary testosterone
remaining in the sex-reversed fish is insignificant in comparison
with the amount produced naturally by a normal adult male tilapia.
One study demonstrated that the plasma testosterone level of sex-
ually active, untreated males in the presence of females was actually
higher than levels found in same-age, sex-reversed fish in monosex
culture. Considering all these factors, it is doubtful that the low re-
sidual levels of the dietary testosterone found in fish months after sex
reversal would be a health hazard for consumers. The use of meth-
yltestosterone for sex reversal of food fish has not yet been approved
by the U.S. Food and Drug Administration.
Can all species of tilapia be sex reversed?
The procedure described in this publication is based primarily on
results obtained with Oreochromis (Tilapia) niloticus, but sex re-
versal with orally administered methyltestosterone at the same dos-
age levels has been demonstrated with other mouth-brooding tila-
pias, including O. aureus, O. mossambicus, O. hornorum and the
"red" tilapia which is generally a mixed 
strain species. It is possible
that the minimum acceptable initial size of fry should be slightly
smaller for the mossambica and hornorum groups.
Hormonal sex reversal has not yet been successful for the sub-
strate spawning tilapias, such as Tilapia rendalli and T. zillii.
Can You Sex Reverse From Male to Female?
After sex reversal with androgens was demonstrated to be feasi-
ble, scientists began considering the use of estrogens to produce
phenotypic female tilapia from genetic males. The fattening of an all-
[141
female population has little appeal.to practical aquaculturists be-
cause females grow more slowly and several could spawn with a sin-
gle accidentally introduced male. However, for some species of
tilapia there existed the possibility of crossing a sex-reversed ho-
mogametic "female" with a normal male of the same species to pro-
duce 100% male offspring. This would not be possible with Oreo-
chromis (Tilapia) niloticus or O. mossambicus but would
theoretically be feasible with O. aureus. With other species of tila-
pia sex reversal with estrogens may make it possible to produce "su-
permales" which produce only male offspring. Unfortunately, sex re-
versal to female with estrogens was much more difficult than sex
reversal to male with androgens. Even with antiandrogens, such as
methalibure, few sex-reversed phenotypic females are produced.
Additional research is needed to clarify the value of estrogens in the
production of all male seed.
SUGGESTED READING
General
YAMAMOTO, T 1958. Artificial induction of functional sex reversal in
genotypic females of medaka (Oryzias latipes). J. Exp. Zool.
132:227-264.
CLEMENS, H.P AND T INSLEE. 1968. The production of unisexual
broods of Tilapia mossambica sex reversed with methyltestos-
terone. Trans. Am. Fish. Soc. 97:18-21.
GUERRERO, R. D. AND W L. SHELTON. 1974. An aceto-carmine
squash method for sexing juvenile fishes. Prog. Fish Cult. 36:56.
GUERRERO, R. D. 1975. Use of androgens for the production of all-
male Tilapia aurea (Steindachner). Trans. Am. Fish. Soc.
104:342-348.
SHELTON, WL., K.D. HOPKINS, AND G.L. JENSEN. 1978. Use of
hormones to produce monosex tilapia. Pages 10-33, R.O0. Smith-
erman, W L. Shelton, J. H. Grover, eds. Culture of exotic fishes
symposium proceedings. Fish Culture Section, American Fish-
eries Society, Auburn, Alabama.
WOHLFARTH, G.W. AND G. HULATA. 1983. Applied genetics of ti-
lapias. ICLARM Studies and Reviews 6. International Center for
Living Aquatic Resources Management, Manila, Philippines.
BYE, VJ. AND R.E LINCOLN. 1986. Commercial methods for the
control of sexual maturation in rainbow trout (Salmo gairdneri
R), Aquaculture, 57:299-309. (Regarding the use of iso-propyl al-
cohol to dissolve the hormone.)
ALVENDIA-CASAUAY, A. AND V.S. CARINO. 1988. Gonadal sex dif-
ferentiation in Oreochromis niloticus. Pages 121-124. R. S.V. Pul-
lin, T. Bhukaswan, K. Tonguthai and J.L. Maclean, eds. The
Second International Symposium on Tilapia in Aquaculture.
ICLARM Conference Proceedings 15. Department of Fisheries,
Bangkok, Thailand, and International Center for Living Aquatic
Resources Management, Manila, Philippines.
Jo, J.Y., R.O. SMITHERMAN, AND L.L. BEHRENDS. 1988. Effects of
dietary 17a- methyltestosterone on sex reversal and growth of Or-
eochromis aureus. Pages 203-207. R.S.V. Pullin, T. Bhukaswan,
K. Tonguthai and J. L. Maclean, eds. The Second International
Symposium on Tilapia in Aquaculture. ICLARM Conference
Proceedings 15. Department of Fisheries, Bangkok, Thailand,
and International Center for Living Aquatic Resources Manage-
ment, Manila, Philippines.
PANDIAN, T.J. AND K. VARADARAJ. 1988. Techniques for producing
all-male and all-triploid Oreochromis mossambicus. Pages 243-
249. R.S.V. Pullin, T. Bhukaswan, K. Tonguthai, and J.L. Ma-
clean, eds. The Second International Symposium on Tilapia in
Aquaculture. ICLARM Conference Proceedings 15. Depart-
ment of Fisheries, Bangkok, Thailand, and International Center
for Living Aquatic Resources Management, Manila, Philippines.
Fry Production For Sex Reversal
HUGHES, D.G. AND L.L. BEHRENDS. 1983. Mass production of Ti-
lapia nilotica seed in suspended net enclosures. Pages 394-401.
L. Fishelson and Z. Yaron, eds. Proceedings International
Symposium on Tilapia in Aquaculture, Nazareth, Israel, 8-13
May 1983. Tel Aviv University, Israel.
ROTHBARD, S., E. SOLNIK, S. SHABBATH, R. AMADO AND I. GRA-
BIE. 1983. The technology of mass production of hormonally sex-
inversed all-male tilapias. Pages 425-534. L. Fishelson and Z. Ya-
ron, eds. Proceedings International Symposium on Tilapia in
Aquaculture, Nazareth, Israel, 8-13 May 1983. Tel Aviv Univer-
sity, Israel.
BUDDLE, R. 1984. Monosex tilapia fry production. ICLARM News-
letter 7:4-6.
POPMA, T.J. 1987 Freshwater fish culture development project, final
technical report, Auburn University/U. of Florida/ USAID tech-
nical assistance contract.
GREEN, B.W 1988. Honduras freshwater aquaculture Project, final
technical report, USAID contract 522-0168-C-00-8010. Auburn
University
GUERRERO, R.D. III AND L.A. GUERRERO. 1988. Feasibility of
commercial production of sex reversed Nile tilapia fingerlings in
the Philippines. Pages 183-186. R.S.V. Pullin, T Bhukaswan, K.
Tonguthai and J. L. Maclean, eds. Second International Sympo-
sium on Tilapia in Aquaculture. ICLARM Conference Proceed-
ings 15. Department of Fisheries, Bangkok, Thailand, and Inter-
national Center for Living Aquatic Resources Management,
Manila, Philippines.
LovSHIN, L. L. AND H. H. IBRAHIM. 1988. Effects ofbroodstock ex-
change on Oreochromis niloticus egg and fry production in net
enclosures. Pages 231-236. R.S.V. Pullin, T. Bhukaswan, K. Ton-
guthai and J.L. Maclean, eds. Second International Symposium
on Tilapia in Aquaculture. ICLARM Conference Proceedings
15. Department of Fisheries, Bangkok, Thailand, and Interna-
tional Center for Living Aquatic Resources Management, Ma-
nila, Philippines.
LIT[LE, D.C. 1989. An Evaluation of strategies for production of
Nile tilapia (Oreochromis niloticus L.) fry suitable for hormonal
treatment. Ph.D. Dissertation, University of Stirling, Stirling,
Scotland.
Sex Reversal In Net Enclosures
CHAMBERS, S.A. 1984. Sex reversal of Nile tilapia in the presence of
natural food. M.S. thesis, Auburn University, Alabama.
BUDDLE, C.R. 1984. Androgen-induced sex inversion of Oreo-
chromis (Trewavas) hybrid fry stocked in cages standing in an
earthen pond. Aquaculture 40:233-239.
BUDDLE, R. 1984. Monosex tilapia fry production. ICLARM News-
letter 7:4-6.
POPMA, T.J. 1987. Freshwater fish culture development project, final
technical report, Auburn University/U. of Florida/ USAID tech-
nical assistance contract.
GREEN, B.W. 1988. Honduras freshwater aquaculture project, final
technical report, USAID contract 522-0168-C-00-8010. Auburn
University.
LANDIVAR-ZAMBRANO, J.J. 1989. Determinacion de la frecuencia
optima de alimentacion para la reversion quimica del sexo de Ti-
lapia nilotica (Oreochromis niloticus). Thesis, Escuela Superior
Politecnica del Litoral, Guayaquil, Ecuador.
Elimination Of Hormone From Tissue
JOHNSTONE, R., J.J MACINTOSH, AND R.S. WRIGHT. 1983. Elimi-
nation of orally administered 17a-methyltestosterone by Oreo-
chromis mossambicus (tilapia) and Salmo gairdneri (rainbow
trout) juveniles. Aquaculture 35:249-257.
ROTHBARD, S., E. SOLNIK, S. SHABBATH, R. AMADO, AND I. GRA-
BIE. 1983. The technology of mass production of hormonally sex-
inversed all-male tilapias. Pages 425-434. L. Fishelson and Z. Ya-
ron, eds. Proceedings International Symposium on Tilapia in
Aquaculture, Nazareth, Israel, 8-13 May 1983. Tel Aviv Univer-
sity, Israel.
GOUDIE, A.C. 1984. Tissue distribution and elimination of radiola-
belled methyltestosterone administered in the diet to sexually
undifferentiated and adult tilapia, Tilapia (Oreochromis) aurea.
Ph.D. Dissertation, Auburn University, Alabama.
Sex Reversal To Female
JENSEN, G.L. AND WL. SHELTON. 1979. Effects of estrogens on Ti-
lapia aurea: implications for production of monosex genetic male
tilapia. Aquaculture 16:233-242.
HOPKINS, K.D., WL. SHELTON, AND C.R. ENGLE. 1979. Estrogen
sex reversal of Tilapia aurea. Aquaculture 18:263-268.
OBI, A. AND WL. SHELTON. 1983. Androgen and estrogen sex re-
versal in Tilapia hornorum. Pages 165-173. L. Fishelson and Z.
Yaron, eds. Proceedings International Symposium on Tilapia in
Aquaculture, Nazareth, Israel, 8-13 May 1983. Tel Aviv Univer-
sity, Israel.
SCOTT, A.G., D. PENMAN, J. BEARDMORE, AND D. SKIBINSKI.
1989. The YY supermale in Oreochromis niloticus (L.) and its po-
tential in aquaculture. Aquaculture, 78: 237-251.
[151