a
*1~
A~.Ir,
k
r -
a
'4i~ ? "44
A
4.:
'A
44~
t, %
'4'
-- -- '4
a
S.;
-W-*-
i0le-
a

CONTENTS
EXECUTIVE SUMMARY _______________________ 2
INTRODUCTION 3
Background 3
Research Station Facilities __________________________ 3
RESEARCH ACTIVITIES ______________________ 4
Fingerling Production ________________________ 4
Fertilization Studies ________________________ 9
Combined Fertilization and Feed Studies _______________20
Other Species Studies ______________________23
Miscellaneous Studies _______________________ 26
On-farm Testing of PD/A CRSP Fish Production Systems _______29
ECONOMICS OF HONDURAN PD/A CRSP POND
MANAGEMENT SYSTEMS _____________________35
Treatment Categories _______________________35
Enterprise Budget Analysis ____________________ 37
Sensitivity Analyses ______________________ 42
CONCLUSIONS 45
Project Publications and Presentations _______________ 46
Appendix 48
Information contained herein is available to all persons
regardless of race, color, sex, or national origin.
IExECUTIVI vE SUIMMARY
A quacultural research has been conducted collaboratively in Honduras since 1983 by the
International Center for Aquaculture and Aquatic Environments, Auburn University and the
Direcci6n General de Pesca y Acuicultura, Secretarfa de Recursos Naturales. This research was
carried out at the El Carao National Fish Culture Research Center, Comayagua, Honduras, under the
auspices of the USAID-financed Pond Dynamics/Aquaculture Collaborative Research Support
Program (PD/A CRSP). The goal of PD/A CRSP is to increase tilapia yields by optimizing resource
use in systems based predominantly on natural pond productivity.
Ponds were stocked with male Nile tilapia (Oreochromis niloticus). Pond nutrient inputs
were organic and chemical fertilizers, and supplemental feed either alone or in some combination.
Fish stocking rate was 10,000/ha during the initial five years of work. During this same period,
experiments were repeated during the rainy and dry seasons on the assumption that seasonal differ-
ences would significantly affect pond productivity. However, temperature proved to be the factor
that affected fish growth most, and the cooler period of the year overlapped the rainy and dry sea-
sons. Thereafter, a warm and cool experimental season was used rather than a rainy and dry season.
Differences in fish yield between warm and cool seasons can exceed 25 percent.
Stocking more than 10,000 fish/ha in organically-fertilized ponds resulted in smaller fish and no
greater fish yields. Increasing stocking rate to 20,000/ha resulted in greater yield when organic
fertilizer (chicken litter) was supplemented with nitrogen as urea. In research on the combination of
organic fertilization and supplemental feeds, feed use was more efficient when combined in low
amounts (1.5 percent biomass/d) with fertilizer, or when used beginning the third or fourth month of
grow-out. Higher economic gains with feed over sole use of chicken litter were never realized at
stocking rates less than 20,000 fish/ha. Economic returns from organic fertilizer plus feed were no
greater than returns from organic fertilizer plus nitrogen as urea. Tilapia yields of 3,500 kg/ha in 150
days were obtained in fertilized ponds without feeds. Yields increased to 5,300 kg/ha in 150 days
when supplemental feeds were used, but high feed cost reduced net returns to less than those for
fertilizers alone.
Assuming a market value independent of fish size, manure plus urea was the most profitable
management system. Sixteen pond management strategies resulted in positive economic returns. All
treatments with positive economic returns used stocking rates of at least 20,000 tilapia/ha. Produc-
tion of large tilapia (> 400 g) necessitates the use of formulated feeds, but a higher market value for
large tilapia is required in order for profitabilty of this management system to exceed that of sys-
tems based on organic fertilization plus urea. Large tilapia generally are produced for export mar-
kets, and require more intensive production practices. Tilapia harvested from semi-intensively
managed ponds can supply domestic markets in Central America. Combined use of organic and
chemical fertilizers as nutrient inputs for tilapia ponds requires less capital expenditure than com-
mercial feeds, and therefore are appropriate for small- to medium-scale commercial producers who
supply domestic markets.
2
DIEVIELOPMENT OF SEMII IINTFIENSIVIE AQUlACUIILTFUIIRIE
TIECHINOILOGII lES IN IHI ONIDUIRAS
SUMMARY OF FRESHWATER AQUACUILTURAL
RESEARCH CONDUCTED FROM 1983 TO 11992
BARTHOLOMEW W. GREEN, DAVID R. TEICHERT-CODDINGTON
1
, AND TERRILL R. HANSON
2
IINTRODUCTION
Three consecutive aquacultural projects have been implemented during the period 1983 to 1993;
research results and economic analyses of production systems are summarized in this report.
BACKGROUND
Since 1983, aquacultural research has been con-
ducted in Honduras as a collaborative effort between the
International Center for Aquaculture and Aquatic Environ-
ments (ICAAE), Auburn University, Alabama, USA, and
the General Directorate of Fisheries and Aquaculture, Min-
istry of Natural Resources (MNR), Republic of Honduras.
Research has concentrated on development of freshwater
fingerling and foodfish production systems, and on semi-
intensive marine shrimp production systems.
Early in 1983, implementation of the centrally-
funded Agency for International Development Title XII
Pond Dynamics/Aquaculture Collaborative Research Sup-
port Program (PD/A CRSP) was initiated in Honduras,
Indonesia, Panama, Philippines, Rwanda and Thailand. PD/
A CRSP research was designed to quantify biological,
chemical, and physical processes of management systems
for pond fish culture. The PD/A CRSP was implemented at
the El Carao National Fish Culture Research Center, Gen-
eral Directorate of Fisheries and Aquaculture, Ministry of
Natural Resources, Comayagua, Honduras. Budget reduc-
tions in 1987 resulted in termination of Honduras, Indonesia,
and Philippines PD/A CRSP activities in August 1987.
Panama was selected as the Latin American site for the PD/
A CRSP world wide effort.
USAID/Honduras, in response to a request from
MNR, contracted ICAAE to provide aquacultural technical
assistance for 15 months following the termination of the
PD/A CRSP. The purpose of this project was to continue
applied research on tilapia fingerling and food fish produc-
tion systems appropriate for Honduras, and to provide aquac-
ultural technical assistance to the USAID/Honduras Natural
Resources Management Project.
Political difficulties forced the cessation of the
Panama PD/A CRSP in December 1987. The PD/A CRSP
was re-established at the El Carao station in Spring 1988
largely through the efforts of the ICAAE. Experiments were
initiated in August 1988.
RESEARCH STATION FACILITIES
Freshwater aquacultural research was conducted at
the El Carao National Fish Culture Research Center,
Comayagua. The station, located about 8 km from the city of
Comayagua, is the largest of a series of aquacultural experi-
ment stations operated by the General Directorate of Fisher-
ies and Aquaculture (Direcci6n General de Pesca y
Acuicultura), Ministry of Natural Resources (Secretarfa de
Recursos Naturales) (Figures 1-2). The El Carao Station is
used principally for the production of tilapia, grass carp,
silver carp, tambaquf, and guapote tigre fingerlings for fish
farmers. Other activities undertaken by the station's techni-
cal staff are extension and some research.
Facilities at the station include offices, a chemical
- biological limnology laboratory, a modest technical li-
brary, and equipment/supply storage areas (Figure 3). Of the
36 ponds, 12 are 0.5-ha, 12 are 0.1-ha, and 12 are 0.2-ha.
Each pond is equipped with a concrete harvest sump. Water
(total alkalinity of 43 mg/L CaCO
3
, total hardness of 31 mg/
L CaCO
3
) is supplied by gravity to ponds from a 0.4-ha
reservoir, which is supplied with water from irrigation
canals originating at the Selguapa River. Pond inlets were
equipped with saran screen filters. A wet-lab area, com-
prised of ten 20-m
2
, four 3-m diameter, and eight 2-m
2
concrete tanks, is supplied with well water. Detailed site
description was given by Egna et al. (1987).
'Green and Teichert-Coddington are Research Fellows in the Department of
Fisheries and Allied Aquacultures, 
2 
Hanson is Senior Research Associate
in the Department of Agricultural Economics and Rural Sociology.
.............. ............................ ..... ... ................... ......................................... ................................................................................ . . ... ...................
..... .......... .......... * ....... * .............. * ....... ....... 
..................
. . ........ ......................................................................... :.:.: .. ...................... ................. .. ............ ..................... ** ....... .
........................... ..... ................................ . .............................. .................................................................................................. ........... ........ ............... x .. ..:.x.::I ................... ...........
..................................................................................................................................................::. ........................................................................................................
....................... ............. . .. ......... .. ............................. ................. . . .
.. .................................................. ...... :..: ......... ....... .............. : ...... I ........... .. ...............
........ ....... 
........
............................................................. ............................... ... .......................................................................... ........................... ............................... . .................................. .... ..........................
.............................. .............................. ................................................................. ............................................. .. ... : ............................ - .: .. ........... . . ....... ....... .. .... . .
................................ ......................................................................................... ........... ......................... .... ................................... 
...................... 
....... 
..............................
............. .......... ... ...... ......... . ................................................
.. ........................ ............................. : ....................................................................................... ...... - ............
.... ..... .. ..... ... ...... ....... 
............. ....... * .......
................................. 
........................................... 
................................................................................................................... 
.......................
............ 
.: .. ......................................................................
* .................. ................................ ................................................................................. ........ .........................................................
................. .................................................................................................................................. ::. . .. ... - .. - .' ::::x'":"*:: ::: ::: .............. ** ....................... ........... ................
.. ..... .. 
............... .........................................
...... ....... *'*'* ... * ..... *" *" * ........................................................................................................... ...................................... ...................... .......................................................................................................................................... ...............
Figure 1. Map of Central America showing location of Honduras and the cities of Tegucigalpa and Comayagua. Figure 2. INSET:
map of Comayagua Valley, Honduras, showing the location of the El Carao National Fish Culture Research Center which is located
at 14026'N latitude and 87041'W longitude, and at an elevation of 583 m above sea level. Not to scale.
RJES IEARCH HIACTIVITIL ES
FINGERLING PRODUCTION
Monosex hybrid tilapia culture has been practiced
in Honduras since at least 1980. Fingerling production
practices at the El Carao station when project implementa-
tion was initiated were based on partial harvests of fingerling
reproduction ponds at 4- to 6-wk intervals. Harvested finger-
lings were stocked into nursery ponds for further growth
prior to distribution to fish farmers. Reproduction ponds
were stocked with female Oreochromis niloticus and male
0. urolepis hornorum. As demand for tilapia fingerlings by
fish farmers increased, it was necessary to intensify produc-
tion practices in order to meet demand. Demand for tilapia
fingerlings continued to increase through the 1980s, but
production of hybrid fingerlings lagged and delays in finger-
ling deliveries resulted.
PD/A CRSP research required the use of male 0.
niloticus. Therefore, fingerling production capability for
this species was established at El Carao in 1983. Hormonal
sex reversal to produce monosex populations was imple-
mented in early 1988 to augment fingerling supplies.
Trials to produce sex reversed 0. niloticus finger-
lings were conducted with fry. produced in earthen ponds.
Prior to pond inundation 1. 27 -cm square mesh knotted nylon
netting was draped over the harvest sump and was held in
place by rocks. A cylindrical drain screen approximately one
meter long and covered with aluminum window screen was
placed in the drain pipe opening to reduce fry loss from
suction during draining. Pond water was lowered until level
with the top edge of the harvest sump, at which time brood
fish were removed from the sump by lifting the nylon
netting. Brood fish were transferred to concrete holding
tanks for restocking the following cycle. Fry were harvested
from the sump with dipnets equipped with 1.6-mm ace nylon
netting.
Water Supply 
Office Laboratory
Canal 
Complex@
RLKRJ Water Tank
RESERVOIR
f Storage
Building
Fish
A A A A A A A A A A A A
Holding
and Tanks
*-Weather
Figure 3. Schematic layout of ponds and support buildings at El Carao National Fish
Culture Research Center. Not to scale.
1 2 3 4 56 7 8 9 10 11 12
B B B B B B
1 2 3 4 5 6 OO0
B B B B B B
7 8 9 10 11 12
- - - - - -Fish
Holding
C C C C C C Building
1 2 3 4 
5 6 and Tanks
- - -
::= Weather
- -- - - -Station
C C C C C C
7 8 9 
10 11 12 
N
Figure 3. Schematic layout of ponds and support buildings at El Carao National Fish
Culture Research Center. Not to scale.
STUDY Al. EFFECT OF HARVEST FREQUENCY ON
HYBRID TILAPIA FINGERLING PRODUCTION
The objective of this study was to determine if more
frequent partial harvests of reproduction ponds would in-
crease hybrid tilapia fingerling production.
Two consecutive studies were conducted in which
female Oreochromis niloticus and male O. urolepis hornorum
were stocked into triplicate 0.05-ha earthen ponds at a rate of
5,000 fish/ha; four females were stocked
per one male. Mean weight of brood 
fish TABLE 1. SUM
EARTHEN I
was approximately 400 g.
Harvest
Ponds were fertilized weekly 
Harvestquency
frequency tc
with chicken litter (140 kg/ha) during the
first study or with fresh dairy cow manure Days
(244 kg/ha) during the second study. Fish 25 4
were fed corn gluten (3 percent of fish 
7 10
El Carao National Fish Culture Research Center
Comayagua, Honduras
per pond at each subsequent partial
harvest decreased with 25-day interval
partial harvests (Figure 5).
Cannibalism of newly hatched fry by larger finger-
lings significantly reduces tilapia fingerling production in
ponds. When partial harvests were conducted at 25-day
intervals, the observed decline in mean fingerling produc-
tion at each subsequent harvest probably resulted from
cannibalism. More frequent harvests (seven-day intervals)
yielded greater cumulative numbers, likely in response to
reduced cannibalism.
IARY OF TILAPIA FINGERLING PRODUCTION (MEAN - SE) IN 0.05-HA
PONDS STOCKED WITH 5,000 BROOD FISH/HA AND SUBJECTED TO
MONTHLY OR WEEKLY PARTIAL HARVESTS
Cumulative
)tal production Daily production Individual weight Duration
No.10.05 ha No./rn
2 /
day g/fingerling days
9,000 + 9,037 0.86 1.5 ? 0.2 114
4,100 ? 17,644 2.19 0.9 ? 0.1 95
biomass) Monday through Friday dur-
ing both studies.
During the first study (15 Janu-
ary to 10 May 1985), fingerlings were
harvested from brood ponds beginning
on day 34 and subsequently at 25-day
intervals, making three passes with a
seine of 6.35-mm square mesh. Ponds
were drained and harvested 114 days
after stocking. In the second study (17
May to 20 August 1985), partial har-
vests of fingerlings were initiated 25
days after stocking and repeated at
seven-day intervals. After 95 days,
ponds were drained and harvested. At
each partial harvest, mean fingerling
weight was determined by weighing
two random samples of 1,000 finger-
lings per pond. Total numbers of fin-
gerlings harvested were calculated from
total weight of fingerlings divided by
mean fingerling weight. Mean water
temperatures in ponds were 24.8 'C
and 27.6 'C during the first and second
studies, respectively.
The total number of fingerlings
harvested increased as the harvest fre-
quency increased (Table 1; Figure 4).
The number of fingerlings obtained
Figure 4. Mean cumulative number of tilapia fingerlings
harvested from 0.5-ha earthen reproduction ponds subjected
to weekly or monthly partially harvests.
STUDY A2. EFFECT OF WATER TEMPERATURE
ON PRODUCTION OF OREOCHROMIS NILOTICUS
FRY FOR SEX REVERSAL
Recently hatched tilapia fry 9 to 11 mm in total
length (TL) are preferred for hormonal sex reversal 
(mascu-
linization) because they are presumed to be sexually undif-
ferentiated. Oral administration of androgen to these fish for
three to four weeks results in populations of fish comprised
of 97 to 100 percent males. Observations made during
preliminary fry production trials indicated that the duration
of the reproduction pond cycle necessary to produce 9 to 11
mm TL fry varied with water temperature. The objective of
this research was to quantify the effect of water temperature,
in terms of degree-days, on Oreochromis niloticus fry pro-
duction for sex reversal in earthen ponds.
Two 0.05-ha earthen ponds were used for each
trial. Thirty-three trials were conducted between 25 Septem-
ber 1988 and 15 March 1990. Oreochromis niloticus brood
fish were randomly allocated to, and concurrently stocked
in, each pond. Mean number of females and males stocked
per pond was 220 and 101, respectively. Average (? 
SD)
individual weights were 233 ? 53 g and 319 ? 39 
g,
respectively. Mean weight of fish by sex was determined
prior to each trial and remained relatively constant across 
all
trials. Broodfish were fed a 23 percent protein pelleted ration
at 3 percent of fish biomass per day, five days per 
week.
In each trial, one pond usually was drained 17 days
after stocking (range 16 to 18 days), and the other 20 days
after stocking (range 19 to 21 days). Fry harvest was sus-
Figure 5. Mean number of tilapia fingerlings harvested from
0.05-ha earthen ponds at weekly or monthly partial harvests.
Mean total fry production
(fry/g female)
3.5 - -i 
Fry not retained 
by grader
3.0 Fry retained 
by grader
2.5
2.0-
1.5
1.0 
n
0.5- 
d
0.0 -..
4.4.5.6.-,-
100 120 140 160 180 200 220 240 260 280
Accumulated degree-days from threshold temperature of 15
0
C
Figure 6. Mean total production of Nile tilapia fry from 0.05-ha
earthen ponds in relation to accumulated degree-days and
shown by 10-degree-day intervals. Number of observations per
10-degree-day interval is given in the box.
pended when fewer than several hundred fry were captured
after two to three consecutive passes with the dipnet. Quan-
tification of the few fry remaining in the harvest sump,
trapped in puddles on the pond bottom, or impaled on 
the
drain screen was not attempted.
Harvested fry were graded through 0.32-cm mesh
vexar which retained fry larger than 13 mm TL. Total
6
Fry retained by 3.2-mm mesh grader
(% of total population)
40 -
35 -
30 -
25-
20-
n
15 o
10- d
a
5
0
180 190 200 210 220 230 240 250 260 270 280
Accumulated degree-days from threshold temperature of 15
0
C
Figure 7. Mean percent of total O. niloticus fry population
retained by 3.2-mm vexar mesh grader in relation to
accumulated 10-degree-day intervals. Average total fry yield
(non-retained plus retained fry) was 86,000 fry/0.05 ha per
harvest.
number of fry per pond not retained by the grader was
determined by visual comparison to a counted sample of
2,000 fry. A random sample averaging 434 fry/pond was
individually measured (TL) to the nearest millimeter.
Pond depth averaged 90 cm. A maximum-mini-
mum thermometer was installed at 50-cm depth in each
pond. Water temperatures were recorded between 0700 and
0800 hours six days per week. Data for missing values was
estimated by averaging water temperature from the preced-
ing and following days. Average daily temperature was
estimated from maximum and minimum daily temperatures.
Calculation of degree-days was based on the difference
between the mean daily temperature and a threshold tem-
perature of 15 'C; the threshold temperature was subtracted
from the average temperature each day, and the results
summed over the trial period.
Daily water temperatures across all trials ranged
from 16.5 to 30.8 oC. Cumulative degree-days per trial
ranged from 119.6 to 276.0 with a mean of 198.2. No fry
were harvested at less than 140 degree-days, but eggs were
observed at the lowest number of degree-days noted. Mean
total fry harvest per gram of female (y) increased signifi-
cantly as cumulative degree-days (x) increased above 140 (y
= 0.021x - 2.605, r
2 
= 0.535, p = 0.0001). Figure 6 shows
mean total fry production per 10-degree-day intervals. Mean
total fry yield per harvest was 86,000 fry per 0.05 ha.
Cumulative fry production during the 33 trials was 4,897,000
fry, 89 percent of which were of the preferred size for
hormonal sex reversal.
Frequency
(% of population)
18.0 -
16.0 -
14.0 -
12.0 -
10.0 -
8.0 -
6.0
4.0
2.0
0.0 
5 
6
30.0 -
25.0 -
20.0 -
15.0 -
10.0 -
5.0 -
0.01
Trial 51-B, 256.2 degree-days
- Fry not retained by grader
Fry retained by grader
7 8 9 10 1112 13 14 15 16 17 18 19 20
Trial 29-A, 167.3 degree-days
6
5 u 11 I
Total length (mm/fry)
13 14
Figure 8. Size distribution of O.niloticus fry population from
two trials to demonstrate effect of high (top) and low (bottom)
cumulative degree-days. Number of fry not retained by the
grader was estimated to be 98,000 and 109,000 fry/0.5 ha for
Trials 51-B and 29-A, respectively. A total of 28,000 fry/0.5 ha
were retained by the grader in Trial 51-B; no fry were retained
in the other trial.
No fry produced between 140 and 195 degree-days
were retained by the grader. Above 195 degree-days, num-
ber of fry retained by the grader (y) increased significantly
with increased cumulative degree-days (x) (y = 0.492x -
98.73, r
2 
- 0.569, p = 0.001; Figure 7).
Between 140 and 280 degree-days the number of
fry not retained by the grader was not linearly dependent on
cumulative degree-days. However, mean total fry harvest
per gram of female was greater between 200 and 260 degree-
days than between 140 and 200 degree-days. Two trials in
the 270 to 280-degree-day interval resulted in low produc-
tion, possibly because temperatures were too high. Fry not
retained by the grader averaged (? SE) 9.5 ? 0.1 mm total
length (TL) and fry retained by the grader were 14.2 ? 0.16
mm TL (Figure 8).
Total fry production and quantity of fry not re-
tained by the grader were consistently higher between 195
and 260 degree-days. The percentage of fry retained by the
grader (oversized fry) nearly doubled between 210 to 220
and 220 to 230 degree-days intervals. Therefore, it appears
that the optimum cumulative degree-days for production of
fry not retained on a 3.2-mm mesh (those suitable for
hormonal sex reversal) and for efficient pond usage would
be between 195 and 220 degree-days.
STUDY A3. EFFECTS OF FRY STOCKING RATE
AND HORMONE TREATMENT DURATION
ON PRODUCTION OF SEX-REVERSED
OREOCHROMIS NILOTICUS
Sex reversal of O. niloticus fry at El Carao has been
practiced in 1.6-mm ace nylon mesh hapas suspended in
ponds and in outdoor cement tanks. Fry are fed a finely-
ground hormone-treated ration (23 percent protein; 60 mg
17-x methyltestosterone (MT)/kg feed) daily in four feedings.
Stocking rate of fry averages 4,300 fry per m
2 
of water
surface area (range: 2,000-5,200 fry/m
2
). Hormone treat-
ment generally lasts 28 days, and has resulted in 98 percent
phenotypic males.
In Honduras, environmental conditions permit year-
round operation although seasonal temperature differences
may affect the efficacy of hormone treatment. The purpose
of the following experiment was to determine effects of fry
stocking rate and hormone treatment duration on sex-rever-
sal of Oreochromis niloticus in hapas suspended in ponds
influenced by seasonal temperature changes.
A completely randomized design in 2 x 3 factorial
arrangement was used. Treatments tested were fry stocking
rate (2,000, 4,000 or 6,000 fry/m2) and hormone treatment
duration (21 and 28 days). Each treatment was replicated
twice. Uniform-age, 9- to 11-mm total length fry were
stocked into hapas (1 m x 1 m x 1 m; 1.6-mm ace nylon
mesh). Number of fry was determined by visual comparison
to a counted sample of 1,000 fry. MT (60 mg/kg feed) was
incorporated into ground, sifted (560- mesh sieve), com-
mercially available feed (23 percent protein). Feeding rates
were 20 percent of biomass per day during week 1, 15
percent of biomass per day during week 2, 12 percent of
biomass per day during week 3, and 10 percent of biomass
per day during week 4. Daily ration was divided into four
equal meals offered at two-hour intervals beginning at 8 a.m.
Fry in replicate hapas within stocking rates were fed the
same quantity of ration daily. Upon completion of the
treatment period, hapas were completely harvested. A ran-
dom sample of 500 to 1,000 treated fry were returned to their
hapa for growth to about five cm, at which time gonads from
about 200 fish were examined microscopically to determine
sex ratio using the aceto-carmine squash method (Guerrero
and Shelton, 1974). Maximum-minimum water tempera-
tures were determined at 0.5-m depth six days/week. The
trials were repeated seven times.
Sex reversal was not significantly affected by dura-
tion of hormone treatment nor by stocking rate, and averaged
100 percent males in all treatments (Table 2). Survival was
variable, but did not differ significantly between treatment
durations or among stocking rates.
There was no significant relationship between fry
growth (g per day) and survival in any experiment. Final fry
weight and length were inversely related to stocking rate (r
2
= 0.615 and r
2 
= 0.581, respectively). Mean water tempera-
ture ranged from 23.5 to 28.5 'C during the seven trials.
Efficacy of hormone treatment was not affected signifi-
cantly in this temperature range.
TABLE 2. EFFECT OF STOCKING RATE AND HORMONE-
TREATMENT DURATION ON GROWTH AND EFFICACY OF
METHYLTESTOSTERONE (MT) TREATMENT FOR SEX REVERSAL
MT treatment Stocking Final Final
duration rate weight length Efficacy
Days Fry/m
2  
g/fry mm/fry Pct. Males
21 2,000 0.13 19.2 99.9
28 2,000 0.21 22.6 100.0
21 4,000 0.10 17.9 100.0
28 4,000 0.13 19.6 100.0
21 6,000 0.08 16.6 100.0
28 6,000 0.11 17.9 100.0
STUDY A4. REPRODUCTION OF GUAPOTE TIGRE
(CICHLASOMA MANAGUENSE):
EFFECTS OF MALE:FEMALE STOCKING
RATIO AND TEMPERATURE
Production of tilapia populations comprised of >
98 percent males has been consistently achieved using sex
reversal technology. However, production of offspring by
unwanted females during grow-out can be significant, and
may have a negative impact on yield of marketable tilapia.
Since the late 1980s, El Carao staff have been recommend-
ing use of guapote tigre (Cichlasoma managuense; 500/ha)
a piscivorous cichlid native to Honduras, to control tilapia
reproduction.
Guapote tigre is a nest breeder that can begin
reproducing in ponds at a few months of age. Number of fry
produced per female appears to be high; however, yields
from reproduction ponds often are low probably because of
predation on larvae by parents or older fingerlings. Repro-
duction may also be affected by stocking rates, ratios of male
and females, and by environmental variables such as tem-
perature. The objective of this study was to document effects
of female:male broodstock ratios and temperature on the
number of fry harvested from reproduction ponds.
Twelve trials were conducted between 28 Septem-
ber 90 and 26 August 91. Two 0.05-ha earthen ponds were
used for each trial. Two treatments (1 F: 1 M and 3 F: 1 M)
were randomly assigned to ponds in each trial. Guapote were
stocked in each pond on the same day. Maximum-
minimum water temperatures were determined at 0.5-m
depth five days per week. Fish were fed a commercial fish or
shrimp ration five days per week at one percent of total adult
biomass. After 25 days, ponds were completely drained and
harvested. Total fry harvests were weighed to the nearest
gram on an electronic balance. A sample of 500 fry was
weighed to determine average weight. Total number har-
vested was calculated as the quotient of harvest weight
divided by average larval weight.
Figure 9. Number of fry harvested (vertical bars) and mean
average water temperature (solid line) during reproduction of
guapote tigre at two female:male ratios from September 1990
to August 1991.
TABLE 3. STOCKING AND HARVEST OF GUAPOTE TIGRE AT
Two FEMALE:MALE RATIOS DURING SEVEN REPRODUCTION
TRIALS IN 0.05-HA EARTHEN PONDS BETWEEN
MARCH 1991 AND AUGUST 1991
Treatment
Variable 1F:1M 3F:1M
Female number ............................. 102 225
Female weight, g ............................ 121 115
M ale number .................................... 102 75
M ale weight, g ................................ 171 164
No. fry .............................................. 27,887 17,077
Fry/m 
2  .. .... ..... ... ..... ...
. ................ 55.8 34.2
Fry/fem ale....................................... 298 80
Fry/gram of female ......................... 2.5 0.7
The first five trials were conducted primarily dur-
ing the cold season. Except for a few hundred fry harvested
during the second month, there was no reproduction until
mean monthly water temperatures rose above 24.5 'C (Fig-
ure 9). The number of fry, fry per female, and fry per gram
of female harvested from the 1F: 1M sex ratio was signifi-
cantly greater than that of the 3F: 1M ratio (Table 3). Lower
harvest numbers at the 3F: 1M ratio were probably related to
greater predation pressure on the fry by parents. After
reproduction started in March 1991, harvest number de-
creased with an increase of female biomass (P < 0.05).
FERTILIZATION STUDIES
All pond production research was conducted in
0.1-ha earthen ponds that had 75-cm average water depth.
Male (manually-selected or sex-reversed) Nile tilapia
(Oreochromis niloticus) fingerlings were stocked during all
experiments. Fish growth was monitored monthly by seine
sample of at least 10 percent of stocked fish. Any tilapia
offspring in the sample was weighed and removed. Reported
gross yield included weight of fingerlings removed during
sampling for the period 1983 to 1988.
At El Carao, the rainy season extended from May
through November, with peak rainfall from June to Septem-
ber. Such seasonal differences were thought to affect
pond dynamics and fish yields, and therefore experi-
ments for the first four years of the project were repeated
during each season. Initiation and completion of either
season varied from year to year and it was difficult to
execute trials completely during a particular season. Usu-
ally one month of each experiment overlapped into the next
season. Later it became apparent that temperature more
than season affected fish yields at the El Carao site. The
cool season ranged from September through February,
and is characterized by water temperatures in the range of
21 to 25 ?C in September to October and from 16 to 21 ?C
from November through February. During the warm season
Resea-ch statij& yovoker.
water temperatures range from 25 to 3 1 'C. T hc s. beginniney
in 1990 experiments were conducted during -he cool and
warm seasons, respectively.
Triple superphosphate (46 percent P.,O-'), urea (46
percent N). diammonium phosphate (18 percent 
N, 46 per-
cent P'05) . dairy cowv manure or laver 
chicken litter were
used as pond nutrient inputs. Chemical fertilizers were
procured locally. Cows manure was obtained 
from the dairy
unit, National Liv estock Center, Ministry of Natural 
Re-
sources, Comayagua:. Fresh manure was scraped from [he
milking parlor beclingc Saturday through Monday 
morn-
;ng when it swas applied to ponds as a rhick slurry. Chicken
'itter, obtained from a local commercial lay er operation, 
wxas
nurchased in bulk for each experiment and stored 
in wxoven
plastic sacks under cov er ani ii it was broadcast over 
the pond
surface. Chicken litter consisted 
of pine sawdust, manure,
feathers and waste Teed. Manure application rates were 
on a
dry matter (DM) bas's. Manure 1DM wxas determined prior 
to
each fertilization.
Water quality analyses were performed according
to methodologies giv en in StLandard Methods (API-A, 
1989)
or Boyd (1979). Water samples were collected with 
a col-
umn sampler (Boyd, 1979); water samples from various
locations within a pond wxere pooled and a subsample 
with-
drawn for analy sis. Primary productixvity wxas determin~ed 
in
ponds using [he free-wxater diurnal curve method. Dissolved
oxygen was measured at tour-hour intervals and 
at 0.25-m
depth intervals using a polarographic dissolved 
oxygen
meter. Measured values were corrected for oxy gen 
diffusion
across the air-water 'Interface using an empirical relationship
that related the oxygen transfer coefficient to xxind 
speed
(Boyd and Teichert-Coddington, 1992).
Economic evaluation of management systems 
tested
is reported in the economic analysis section of this report.
SiTUDY BE. PRODJC1PION OF 'P LAPIA IN 
PONDS
}FhRTIL2ZED WIlTH PH~OSPHORUS ONLY
This wxas the first PD/A CRSP experiment con-
ducted in Honduras. It was designed to establish a baseline
of data during dry and rainy seasons for tilapia 
ponds that
received minimal nutrient inputs.
Ten and 12 ponds were used for the dry and rainy
season study . respectively. Ponds were stocked 
wxith 10,000
tilapia/ha. Mean individual weight of fingerlings was 13.1
g and 10.4 gfor the dry and rainy season studies, 
respec-
tively. Ponds were fertilized with triple superphosphate
(TSP; 46 percent PO 5 8.7 kg TSP/ba) every two 
weeks.
Both studies lasted 150 days.
Mean tilapia gross and net y ield wxere significantly
greater during dry season than rainy season (Table 
4). Mean
individual fish weight at stocking was different for each
season; an analysis of covariance indicated 
no significant
effect of stocking weight on gross yield. No seasonal 
differ-
ence in fish survival was observed. Mean early 
morning
water temperatures were 24.7 'C and 25.7'C 
tor dry and
rainy seasons, respectively.
Mean ne and gross primary productivity wxere
significantly greater during the dry season than during 
the
rainy season. There were acequate concentrations of 
inor-
g7anic nitrogen and phosphorus (Table 5) for high primary
productivity, but photosynthesis apparently was limited 
by
high clay turbidity. Combined tilapia yield (both seasons)
was significantly correlated to net primary productivity (r 2
0.795) (Figure 10'. indicating tha. higher tilapia yields
might hasve been obtained wxith higher rates, of 
primary
production.
800
600
500
300 0 0 Rainv cao
200 [C 
Drx seaIson
0.00 0.50 1.00 1.50 2.00 
2 50
.Net prirn prod ui to Oni ri 
C de
Figure 10. Relationship betwseen mean tilapin yied 
after 150
days and net primary producivly in ponds ierti Dad 
with
phosphorus only during diD ard rairy season experiments.
TABLE 4. PRODUCTION RESULTS FROM 0.1-HA EARTHEN PONDS STOCKED WITH MALE OREOCHROMIS NILOTICUS (10,000/HA) AND
FERTILIZED WITH TRIPLE SUPERPHOSPHATE. TEN PONDS WERE STOCKED DURING THE DRY SEASON AND 12 DURING THE RAINY SEASON
Dry Season Rainy Season
Variable Mean S.E. Mean S.E. t value
Gross yield, kg/ha/iSO days ............................... 547 35.9 334 25.2 -4.84*
Net yield initial stock, kg/ha/iSO days ............... 416 34.6 289 32.9 -2.68*
Tilapia reproduction, kg/ha/iSO days..................... 4 3.0 94 26.5 3-.08*
Survival, pct................................................. 91.1 1.1 90.2 2.0 -0 .41
Final individual weight, g/fish ............................. 60 3.8 37 2.8 -4.84*
*Seasonal means were significantly different (P<0.05).
TABLE 5. SUMMARY OF PRIMARY PRODUCTIVITY AND WATER QUALITY VARIABLE MEANS, BY SEASON, IN 0.1-HA EARTHEN PONDS
STOCKED WITH MALE OREOCHROMIS NILOTICUS (10,000/HA) AND FERTILIZED WITH TRIPLE SUPERPHOSPHATE. TEN PONDS WERE
STOCKED DURING THE DRY SEASON AND 12 DURING THE RAINY SEASON
Dry season Rainy season
Variable Mean S.E. Mean S.E. t value
Gross primary productivity, g 0 
2
/M
3 
/day ............. 1.7 0.17 2.6 0.17 na
Net primary productivity, g 0 
2
/m
3
/day ...................... 1.0 0.13 0.8 0.11 na
Early morning pH ........................................... 8.32 9.47 8.14 9.31 4.11*
Total alkalinity, mg/L CaO
. . . . . . . . .
106.7 5.0 85.5 4.3 -3.21 *
Total hardness, mg/L CaCO
3. . . . . . . . . .
79.0 2.0 62.6 2.8 -4.76*
Ca
2
l hardness, mg/L CaCO
3. . . . . . . . . .
63.0 1.2 51.0 2.5 -434*
Ammonia, mg/L NH 
3
-N .................................... 0.23 0.03 0.38 0.02 3.84
Nitrate, mg/L N0 
3
-N ........................................ 1.27 0.06 1.49 0.04 2.99*
Soluble orthophosphate, mg/L P0 
4
-P .................. 1.02 0.06 0.87 0.04 -2.14*
na Statistical test not appropriate.
* Seasonal means were significantly different (P<0.05).
Based on reports in the literature for phosphorus-
fertilized fish ponds, fish yields were expected to average
1,000 kg/ha. The principal factor limiting fish production in
the present experiments probably was low primary produc-
tivity that resulted from light-limitation caused by clay
turbidity. Although incoming water generally had a slightly
milky color, water appeared to be more turbid during the
rainy season as a result of surface runoff. However, much of
the turbidity probably was caused by the suspension of fine
clay particles from the bottom muds by wind-induced water
circulation.
STUDY B2. TILAPIA YIELD IN PONDS
FERTILIZED WITH SIMILAR QUANTITIES OF
NITROGEN AND PHOSPHORUS AS ORGANIC
OR CHEMICAL FERTILIZER
Fertilization with phosphorus resulted in low tila-
pia yield (Study B 1). High levels of clay turbidity in ponds
apparently limited primary productivity and fish production.
Organic fertilizers stimulate primary and secondary produc-
tivity in ponds, and also can precipitate clay turbidity.
Higher quantities of inorganic N and P inputs also could
stimulate primary production and fish yields. The objective
of this study was to test the hypothesis that organic and
chemical fertilization, based on similar N and P inputs, result
in the same fish production.
This study was repeated during the dry and rainy
seasons. Dry season treatments (four replicate s/treatment)
were layer chicken litter, dairy cow manure, and urea plus
triple superphosphate. Rainy season treatments (six repli-
cates per treatment) were chicken litter and urea plus triple
superphosphate. Similar quantities of nitrogen and phospho-
rus were applied in each treatment except for cow manure
which had a high N:P ratio (Table 6). Chemical fertilizer was
not mixed with organic fertilizer to equalize N and P inputs.
Tilapia fingerlings were stocked into ponds at a rate
of 10,000/ha. Fingerlings averaged 33 g and 17 g during the
dry and rainy seasons, respectively. Dry and rainy season
experiments were initiated on 16 January 1985 and 26 July
1985, respectively. Duration of each experiment was 150
days.
During the dry season, mean tilapia gross yield was
significantly greater in the chicken litter treatment than in the
11
TABLE 6. SUMMARY OF NITROGEN, PHOSPHORUS AND POTASSIUM CONTENTS (PERCENT DRY MATTER [DM]
OF NUTRIENT SOURCES AND NUTRIENT APPLICATION RATES DURING DRY AND RAIN SEASON FERTILIZER
EXPERIMENTS IN PONDS STOCKED WITH NILE TILAPIA (10,000/HA)
Total application, kg/ha
Nutrient source N - P - K
1  
Application rate Nitrogen Phosphorus
Pct. kg DM/ha/week
Dry Season
Layer chicken litter, 83.3% DM ... 2.75 - 2.46 -2.33 500 302 270
Dairy cow manure, 21.3% DM ..... 1.46 - 0.55 -0.70 1,020 328 123
Urea ............................................. 46.00 - 0.00 -0.00 
30.6 295 -
Triple superphosphate
2 
.....
.... . . . ....  
0.00 -20.10 -0.00 62.6 - 264
Rainy Season
Layer chicken litter, 84.5% DM ... 2.48 - 1.70 -2.53 500 260 179
Urea ............................................. 46.00 - 0.00 -0.00 
31.3 302 -
Triple superphosphate.................. 0.00 -20.10 -0.00 49.3 - 238
1 N - P - K: Nitrogen - Phosphorus - Potassium content, dry matter basis.
246.00 
percent 
P205.
TABLE 7. PRODUCTION DATA (MEAN ? S.E.) FROM 0.1-HA EARTHEN PONDS STOCKED WITH MALE
OREOCHROMIS NILOTICUS (10,000/HA) THAT RECEIVED ORGANIC OR CHEMICAL
FERTILIZER DURING THE DRY AND RAINY SEASONS
Gross Net yield of Net total
Treatment Final weight Survival yield initial stock yield
g/fish Pct. kg/ha/150 days
Dry Season
Layer chicken litter ...... 204 ? 8.1 a 96.9 a 2,075 ? 89 a 1,663 ? 65 a 1,759 ? 88 a
Dairy cow manure ........ 172 ? 3.8 b 95.7 a 1,626 ? 39 b 1,272 ? 50 b 1,295 ? 45 b
Chemical fertilizer ........ 150 ? 9.0 b 93.6 a 1,513 +106 b 1,071 ? 96 b 1,194 ?110 b
Rainy Season
Layer chicken litter ...... 183 ? 12.0c 93.0c 1,594? 72c 1,426? 71 c 1,530+ 84c
Chemical fertilizer ........ 132 ? 10.0 d 92.9 c 1,153 ? 95 d 987 ? 96 d 1,301 +114 d
ab and cd Columns means within season followed by the same letter were not significantly different
(P>0.05).
TABLE 8. COMPARISON OF DRY AND RAINY SEASON, AND TREATMENT YIELDS OF NILE
TILAPIA (OREOCHROMIS NILOTICUS; 10,000/HA) IN 0.1-HA PONDS FERTILIZED WITH
CHICKEN LITTER OR CHEMICAL FERTILIZER
Variable Mean weight 
Gross Net yield of Total 
net
Initial Final Survival yield intital stock 
yield
g/fish Pct. kg/ha/150 days
Chicken Litter
Dry season .................. 34 204 96.9 2,075 1,663 1,759
Rainy season............. 17 160 93.0 1,594 1,426 1,531
Chemical Fertilizer
Dry season ................... 33 150 93.7 1,513 1,071 1,194
Rainy season............ 17 119 92.9 1,153 987 1,032
Treatment Means
Chicken litter............. 23 178* 94.5 1,786* 1,521* 1,622*
Chemical fertilizer ........ 23 132 93.2 1,297 1,021 1,097
Seasonal Means
Dry seasonn................. 33* 177* 95.3 1,794* 1,367 1,477
Rainy season................ 17 140 92.9 1,373 1,207 1,281
* Means were significantly different (P<0.05).
other treatments, but
dairy cow manure and
chemical fertilizer
treatment yields
were not signifi-
cantly different
(Table 7). During the
rainy season, tilapia
yield was signifi-
cantly greater in the
chicken litter treat-
ment than in the
chemical fertilizer
treatment (Table 7).
Chicken litter appli-
cations also resulted
in significantly
greater tilapia yield
than did chemical
fertilizer applications
when dry and rainy
season data were
pooled (Table 8).
Tilapia net yield did
not differ signifi-
cantly between sea-
sons (Table 8). Mean
early morning water
temperatures were
24.5'C and 25.7'C
for dry and rainy sea-
sons, respectively.
Tilapia yields
were positively correlated to
primary productivity. Appli-
cations of chicken litter and
chemical fertilizer resulted in
higher primary productivity
than did application of cow
manure (Table 9). However,
chicken litter resulted in sig-
nificantly greater primary
production than chemical fer-
tilizer (Table 10). Greater bio-
logical oxygen demand, as
indicated by community res-
piration, resultedfrom organic
fertilizer (Tables 9 and 10).
Organic matter helped clear
clay turbidity that was ap-
parently limiting photosyn-
thesis, because gross primary
12
TABLE 9. MEAN PRIMARY PRODUCTIVITY (G 02/M
3 
PER DAY) AND
COMMUNITY RESPIRATION (G 02/M
3 
PER DAY) IN PONDS
FERTILIZED WITH CHICKEN LITTER OR CHEMICAL FERTILIZER
DURING THE DRY AND RAINY SEASON EXPERIMENTS
Primary productivity
Community
Season Net Gross respiration
Dry Season
Layer chicken litter 1.92 ? 0.39 a 4.53 ? 0.79 a 5.36 ? 0.88 a
Dairy cow manure 0.77 ? 0.21 b 2.93 ? 0.44 b 4.20 ? 1.19 ab
Chemical fertilizer 1.84 ? 0.67 a 3.76 ? 1.08 ab 3.83 ? 2.05 b
Rainy Season
Layer chicken litter 2.52 ? 0.20 c 7.44 ? 0.44 c 9.84 ? 0.48 c
Chemical fertilizer 1.61 ? 0.23 d 4.72 ? 0.64 d 6.21 ? 0.84 d
ab and cd Column means within season followed by the same letter were not
significantly different (P <0.05).
increases as pH rises. Large application of ammonia-based
fertilizer may cause un-ionized ammonia concentration to
threaten fish survival.
Dairy cow manure and chemical fertilizer were
equally effective nutrient sources in tilapia grow-out ponds,
while chicken litter was the most productive. Tilapia yield
for the chemical fertilizer treatment was greater than that
observed with phosphorus-only fertilization (Study B 1);
however, greater quantities of phosphorus, in addition to
nitrogen, were added to ponds in the present study. No
turbidity control was employed when phosphorus was the
only nutrient added to ponds.
STUDY B3. TILAPIA PRODUCTION IN RESPONSE TO
CHICKEN LITTER FERTILIZATION
Study B2 demonstrated chicken litter resulted in
TABLE 11. COMPARISONS OF WATER QUALITY MEANS IN PONDS FERTILIZED
WITH CHEMICAL FERTILIZER OR CHICKEN LITTER
Total Total Soluble
alkalinity Ammonia phosphorus orthophosphate
Treatment pH (mg/L as CaCO3) (mg/L NH
3
-N) (mg/L PO
4
-O)
Chicken litter
Dry season ........... 7.8 131.3 0.25 5.5 4.0
Rainy season......... 7.8 137.7 0.58 5.1 3.8
Chemical fertilizer
Dry season ........... 8.5 66.3 0.56 9.5 6.5
Rainy season......... 7.9 64.4 0.67 13.8 10.1
Treatment means
Chicken litter ........ 7.9* 135.2* 0.45* 5.3* 3.9*
Chemical fertilizer 8.0 65.1 0.62 12.1 8.7
Seasonal means
Dry season ........... 8.0 98.8 0.40* 7.5 5.2
Rainy season......... 7.9 101.0 0.62 9.5 6.9
*Means within heading are significantly different 
(P<0.05).
high fish yield. There were
no known studies that sys-
tematically related tilapia
yields to chicken litter in-
put. Such data would be im-
mediately applicable to the
host country and provide a
basis for economic analy-
ses. The objective of this
study was to quantify tila-
pia yields in ponds fertil-
ized with chicken litter.
Layer chicken litter
(CL) was applied to ponds
weekly at rates of: 125 kg,
250 kg, 500 kg or 1,000 kg
13
productivity and Secchi disk visibility in-
creased together.
Fertilizer type resulted in significant dif-
ferences in water quality. Mean total alkalin-
ity was significantly greater and mean pH,
total ammonia nitrogen (TAN) and phospho-
rus were significantly lower in the chicken
litter treatment than in the chemical fertilizer
treatment (Table 11). Calcium carbonate
present in the chicken litter and pond soils
was dissolved by carbon dioxide generated
by bacterial decomposition of chicken litter,
thus increasing total alkalinity. Ammonia is a
base which can increase pH; the proportion of
un-ionized ammonia, which is toxic to fish,
TABLE 10. SEASONAL COMPARISON OF PRIMARY
PRODUCTIVITY (G 0/M
3 
PER DAY) AND COMMUNITY
RESPIRATION (G O
2
/M PER DAY) IN PONDS FERTILIZED
WITH CHICKEN LITTER OR CHEMICAL FERTILIZER
Primary productivity
Community
Variable Net Gross respiration
Chicken litter
Dry season .............. 1.92 4.53 5.36
Rainy season............ 2.52 7.44 9.84
Chemical fertilizer
Dry season ............... 1.84 3.76 3.83
Rainy season............ 1.61 4.72 6.21
Treatment means
Chicken litter .......... 2.28* 6.28* 8.05*
Chemical fertilizer... 1.71 4.33 5.25
Seasonal means
Dry season .............. 1.88 4.15* 4.60*
Rainy season........... 2.07 6.08 8.03
* Means are significantly different (P<0.05)
Gross tilapia yield
(kg/ha per 150 d)
2,600 r-
y = 864.644 + 2.804x - 0.00lx2
- r
2 
= 0.897 .--
O0
0
O
I I i I I ! I I
2,400
2,200
2,000
1,800
1,600
1,400
1,200
1,000
800
10(0
Figure 11. Nile tilapia production function in relationship to
chicken litter input in 0.1-ha earthen ponds in Honduras.
Rainy and dry season data are combined.
DM/ha. The study was repeated during the rainy and dry
seasons. Three 0.1-ha earthen ponds were randomly as-
signed to each treatment during each season. Nile tilapia
were stocked at 10,000/ha. Mean fingerling weight (g/fish)
was 26.1 and 36.6 during the rainy and dry season, respec-
TABLE 12. SEASONAL COMPARISONS OF NILE TILAPIA PRODUCTION I
(MEAN ? S.E.) FROM 0.1-HA EARTHEN PONDS FERTILIZED WEEKLY
CHICKEN LITTER (CL) ON A DRY MATTER BASIS
Variable Rainy season Dry season t-
125 kg CL/ha per week
Gross yield, kg/ha 1,179 ? 162 1,145 ? 52 0.
Reproduction, kg/ha 61 ? 12 43 ? 25 0.(
Net yield, kg/ha 915 ? 164 781 ? 53 0.
Survival, pct. 94.1 ? 0.6 92.9? 0.9 1.
Individual weight, g/fish 115 ? 15.4 117 ? 6.4 -0.
250 kg CL/ha per week
Gross yield, kg/ha 1,649 + 112 1,426 ? 122 1.
Reproduction, kg/ha 158 ? 10 35 ? 
18 5.
Net yield, kg/ha 1,381 ? 117 1,050 ? 111 2.
Survival, pct. 92.7 ? 2.8 96.7 ? 3.3 -0.
Individual weight, g/fish 155 ? 10.0 143 ? 9.7 0.
500 kg CL/ha per week
Gross yield, kg/ha 1,890 ? 60 1,915 ? 45 -0.
Reproduction, kg/ha 185 ? 21 41 ? 21 4.
Net yield, kg/ha 1,643 ? 76 1,543 ? 43 1.
Survival, pct. 93.0 ? 1.0 91.1 ?0.8 1.
Individual weight, g/fish 177 ? 3.4 203 ? 4.8 -4.
1,000 kg CL/ha per week
Gross yield, kg/ha 2,324 ? 70 2,333 ? 15 -0.
Reproduction, kg/ha 221 ? 68 51 ? 43 2.
Net yield, kg/ha 2,046 ? 76 1,964 ? 17 1.(
Survival, pct. 96.6 + 0.8 84.8 ? 0.6 11.
Individual weight, g/fish 209 + 7.6 268 ? 7.1 -5.(
**Seasonal means were significantly different (P<0.01).
300 500 700 900 1,100
Chicken litter (kg dry matter/ha per wk)
increased (Table 14). Season did not significantly
affect water quality variables. Mean dissolved in-
14
, I I - - - I - - 1
Gross fish yield
(kg/ha per 150 d)
2,600 - y = 
623.328 + 119.667x
2,400 r
2 = 
0.828 00
0 0
2,200 0
2,000 - O O
1,800 -
O0
1,600 -
1,400 
O 
0
1,200 0 
Q0
1,000 0
{] I I I I I I I
002 4 6 8 10 12 14 16
Gross primary productivity (mg 02/L per day)
Figure 12. Relationship between gross tilapia yield and gross
primary productivity in 0.1-ha earthen ponds fertilized weekly
with layer chicken litter. Data from rainy and dry seasons.
tively. Rainy and dry season experiments were started on 5
June 1986 and 7 February 1987, respectively, and lasted
150 and 152 days, respectively.
Tilapia yield increased from 934 to 2,451 kg/ ha,
and from 1,085 to 2,363 kg/ha during the rainy and dry
seasons, respectively, as weekly CL input increased
DATA from 125 to 1,000 kg DM/ha. Gross tilapia yield (y)
WITH increased curvilinearly 
as manure application 
rate
(x) increased (y = 864.6 + 
2.80x - 0.001x
2
, r
2 
=
Value 0.897) (Figure 11). No seasonal differences in
gross or net tilapia yield were observed (Table 12).
198 Greater quantities 
of tilapia offspring were har-
639 vested during the rainy season, indicating that the
774 larger the fingerling 
at sexing, the more effective
110
130 the separation 
of sexes. Tilapia yield 
was positively
correlated with netprimary production (r
2
= 0.828),
348 gross primary production (Figure 
12), and chloro-
963**
052 phyll a concentration 
(r
2 
= 0.836).
912
858 Mean net primary production ranged from
1.15 to 6.45 and from 1.49 
to 7.90 g 0
2
/m
3 
per day338
887** during the rainy and dry 
seasons, respectively. Net
157 and gross primary production and community res-
408 piration all increased significantly with increasing
352**
rates of fertilization. Dry-season 
net primary pro-
124 duction for the 500 and 1,000 kg/ha per week
127 treatments was greater than that for the rainy sea-
063
620** son (Table 13). Water quality 
concentrations gen-
690** erally increased as chicken litter application rate
I
TABLE 13. SEASONAL COMPARISON OF PRIMARY PRODUCTIVITY AND
COMMUNITY RESPIRATION TREATMENT MEANS G 0
2
/M
3 
PER DAY; + S.E.) IN
0.1-HA EARTHEN PONDS STOCKED WITH OREOCHROMIS NILOTICUS AND
FERTILIZED WITH CHICKEN LITTER (CL)
Variable Rainy season Dry season t-Value
125 kg CL/ha per week
Community respiration 5.68 ?0.79 6.66 ?0.67 -0.96
Net primary productivity 1.46 ?0.21 2.29 ?0.27 -2.44
Gross primary productivity 4.31 ?0.55 5.62 ?0.61 -1.60
250 kg CL/ha per week
Community respiration 9.49 ?1.39 8.29 ?1.16 0.66
Net primary productivity 3.31 ?0.66 2.97 ?0.74 0.34
Gross primary productivity 8.04 ?1.28 7.11 ?1.29 0.52
500 kg CL/ha per week
Community respiration 11.60 ?0.31 12.11 ?0.41 -1.01
Net primary productivity 4.03 ?0.32 5.51 ?0.31 -3.32*
Gross primary productivity 9.83 ?0.47 11.57 ?0.51 -2.52
1000 kg CL/ha per week
Community respiration 13.92 ?1.06 13.93 ?0.65 -0.01
Net primary productivity 5.37 ?0.61 7.44 ?0.24 -3.19*
Gross primary productivity 12.33 ?1.12 14.40 ?0.57 -1.66
*Seasonal means were significantly different (P<0.05)
STUDY B4. TILAPIA YIELD IN PONDS
STOCKED WITH 20,000 FISH/HA
AND FERTILIZED WEEKLY WITH CHICKEN
LITTER (1,000 KG DM/HA)
Previous fertilization trials were conducted
at a fish stocking rate of 10,000/ha. However, many
Honduran fish farmers routinely stock fertilized
ponds at 20,000 fish/ha. Results of Study B3 dem-
onstrated that the highest tilapia yield was obtained
in ponds fertilized with chicken litter weekly at
1,000 kg DM/ha. The objective of this study was to
ascertain the effect of doubling stocking rate on
fish yield and average individual weight in ponds
fertilized weekly with chicken litter at 1,000 kg
DM/ha.
On 17 July 1988, three ponds were stocked
with tilapia fingerlings (16.5 g/fish) at a rate of
organic nitrogen concentration was low which indicated that
primary productivity may have been nitrogen-limited. In-
deed, Studies B5 and B6 demonstrated that primary produc-
tion and fish yields could be increased by supplemental
nitrogen fertilization.
Study B2 demonstrated that chicken litter, applied
at 500 kg/ha per week, resulted in greater fish yield than fresh
cow manure or high doses of chemical fertilizer alone. This
study demonstrated that tilapia yield was increased further
through greater applications of chicken litter. At the highest
rate of CL fertilization, 1,000 kg DM/ha per week (equiva-
lent to 143 kg DM/ha per day) were being applied to the
pond.
20,000/ha. Ponds were fertilized weekly with layer chicken
litter at 1,000 kg DM/ha. Ponds were harvested after 150
days.
At harvest, mean gross yield was 2,203 kg/ha,
while the mean net yield was 1,873 kg/ha. Fish survival
averaged 92 percent. Mean gross yield was similar to that
obtained during Study B3 (2,262 kg/ha). The effect of
stocking rate on final fish size was clear: mean individual
weight at 20,000/ha was 120 g/fish, half that obtained at
10,000 fish/ha. Fish harvested from this experiment were
barely of marketable size. A longer production cycle prob-
ably would have demonstrated even greater growth depres-
sion at the higher density. In maunured ponds, the lower
stocking rate appears advisable for commercial production.
15
TABLE 14. MEANS OF WATER QUALITY ANALYSES FOR SAMPLES COLLECTED WEEKLY FROM PONDS STOCKED WITH OREOCHROMIS
NILOTICUS AT 10,000/HA AND FERTILIZED WITH DIFFERENT RATES OF CHICKEN LITTER
Weekly chicken litter application
125 kg/ha 250 kg/ha 500 kg/ha 1,000 kg/ha
Variable Rainy Dry Rainy Dry Rainy Dry Rainy Dry
Soluble orthophosphate, mg/L PO
4
-P 1.73 1.94 3.59 2.99 4.17 4.33 3.79 5.80
Total phosphorus, mg/L PO
4
-P 2.32 2.44 4.53 3.45 5.40 5.33 5.23 7.38
Organic nitrogen, mg/L 1.26 1.33 1.68 1.47 2.07 2.28 2.48 2.74
Ammonia-nitrogen, mg/L NH3-N 0.06 0.05 0.07 0.04 0.07 0.05 0.07 0.09
Nitrate-nitrogen, mg/L NO
3
-N 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01
Nitrate-nitrogen, mg/L NO
2
-N 0.002 0.003 0.002 0.002 0.002 0.002 0.002 0.002
Total alkalinity, mg/L CaCO
3  
101.89 127.58 134.26 126.77 140.25 161.99 160.62 199.70
Chlorophyll a, mg/m
3  
30.58 50.66 69.96 60.82 115.89 112.98 160.27 175.62
Secchi disk visibility, cm 11.3 16.0 15.9 19.6 16.3 17.3 20.8 23.7
6:00 a.m. dissolved oxygen, mg/L 2.62 2.96 1.42 2.39 0.62 1.00 0.30 0.66
6:00 a.m. water temperature, OC 25.4 24.7 25.5 24.9 25.8 24.9 26.0 25.1
SUYB5. SUP~PLEMENTAL N ]7R GEN
-7EW=LATION OF Dm.G N]I ALLY
INEIRTLEZED PONDS
Study B3 suggested that primary productivity and
fish yield in organically fertilized ponds at 
El Caran were
limited by dissolvedl inorganic nitrogen concentratin 
. The
objec'ive of this st, dy was to determioe 
if supplemertalin
of organic -fertilization with ntroger -Fertilizer 
woull in-
crease prim ary productivity and tilaoia 
yield.
Six ponds were stocked 00 " 3 February 
1990 witl
Nile 'ilana at lJ,00/1ta, ard guanote 
t(gre (Cichiasoma
rnaaoguensie) at 250Pha to prey upon any 
tiiapia offspring
produced. Al p nds were fertilized weekly 
with chicken
litter a, 750 kg DIV/ha, ard three ponds also 
received urea
-fertilizer (46-0-0) weekly at 10 kg N/ha. Ponds 
were bar-
vested 126 days afte ci s ckng. Chlorophyll 
a, total ammo-
TABiL5. TIAPI PRODUTIO AN 7/1Ah QUALITY' 
YOAA(M:
FROMi 0.1-H AR HRx PrNDS STOCEau WITH MALE 
OREOCfiROJSA
AN FF'ULZFD 1 0 i CICNe. LITTER (750 
oG oxY VIATER/HiA PER
CHICKEN Ln-z SiFmENTE WITH rio Tr REA (110 
KG N/HA PER Vv
Treatment
Manure only Manure
Variable
Gross yield dkgfna -er 126 L...........
Mean i mdi'ida alvvcight, g/fish .....
Survival, pet.............................
Chlorophyll o, mag/ri..................
Totai ammonia-nirogen. mg/h NH_ I
Filterable orthophosphate, mg/h P0
4
-P
Total phosphorus, mg/I. PG
4
-0.....
1,401 22
166 -- 2.4
84? 1.t
344 + 32.9
0.09 j,0.0 11
2.66+0.118
4.4(0 ?0.274
1,527+
177?
86+
47i
0.16
1.93
3.48
nia nitrogen 'TAN), iota' phosphorus, and Filterable ortho-
phosphate w ere determined on a reg ular basis.
Gross ish yield arid average fish size at ha-est
were not significantly greater in urea-fertili 
zed poods (T.able
15). Mean chlorop1hyllo aid 
T
AN weresignifcantly g-eate
in ponds treatec with urea, whereas 
Lotal phosphorus an1
filterable orthophosphate were significantly 
less (Table 15).
Nitrogen supplemnt ation increased dissolved 
inorganic
nitrogen concentrations and plankton 
biomass. Lower filter-
able orthopnosphate resulted fronm greater 
phosphorus up-
take by pbytoplankton.
STUDYh{~ B6 V ATO OF NUTREN 
I ENIPU (E-uNN
RA7TtO BY SUPPLLMENIHNC 3iRI. 
NLIL
FEWTILIZED :PONDS WITH- N]ITROGEN
Organically fert lized ponds in Honduras 
shave con-
sistently low concentrations (< 0.1 wg/L) 
of total ammonia-
anid nitrate-nitoge i. Low issolved iorganic 
itrogen con-
centrations -lay inicate nit -ogen-lim 
iiation of primary pro-
ductivity. Results of Study B5 suggested 
the possibility for
increasing prmary produictinn and tilapia 
yield by supple-
menting organic fertilizers with niioogen. 
The objective of
this research was to quantify dhe cffect 
of different rates of
itrogen supplementation on -prim ary productivity 
and tila-
pia yield in organically ferdlizeil 
ponds.
Twelve ponds were randomly assigned to fourT
treatments. A il ponds were tfertlized weekly 
with chicken
litter at 750 kg DM/ha. Ir addition, urea 
was applied to
achieve C: ratins of 1 1:1 (corresponds to 
no supplemental
N), 8: 1, 6: 1, or 4: 1. Urea application rates 
were calculated
based on frequent determinations of manure 
nitrogen and
organic carbon. Ponds were stocked on 
26 February 1991
with tilapia fingerlings at 20,000/ha. Cichlasomer
S) mwzaguense fingerlings were stockred 
at 500/ha
7. 110I U to prey upon 
arny tilapia reproduction. 
All ponds
VtEx) OR were harvested 154 days after stocking.
E -K
Vean total nitrogen and organic car-
bon composition of chicken litter was~ 
2.3 ano
plsue 25.0 percent, respeetively. 
Total weekly nitro-
132 gren irputs, including supplemental 
urea, ranged
5.0 from 17.1 kg/ha for control ponds to 46.9 kg/ha
5. for C4:N porids (Table 
t6). Resonse of fis
+ 0.0l6* yield to increased levels 
of supplemental nitro-
+ 0. 168* gen was discontinoous (Table 16) with large
018 variation in som e treatments. Oily 
the C6./
compari- ratio resulted in signilficantly 
increased fish yield
compared with the Control. These. yields 
were
*Treatment neaps wiere significantly different (P<0.05): horizontal
snns only.
substantially higher than previously recorded yields from
fertilized ponds at El Carao. Analyses were complicated by
complete fish mortalities in one replicate pond of each
treatment except the control. The first die-off occurred
during the second month in treatment C6:N1, 12 hours after
ponds were fertilized with urea. A spike in un-ionized
ammonia was suspected, although early morning total am-
monia (< 0.1 mg/L) and pH (8.2) were low two days before
the kill. The weekly urea dose for all treatments subse-
quently was divided into two applications per week. A 95
percent fish mortality in pond B9 (treatment C8:N1) oc-
curred during the last month of the study; fish growth had
been good up to that point. A complete mortality in pond B 10
(treatment C4:N 1) occurred during the last month, following
several months of low growth (Figure 13) apparently in-
duced by a heavy blue-green algae bloom. Mean fish yield
showed a tendency to increase with increasing chlorophyll
Gross fish yield
(kg/ha per 150 d)
225 - A
AC4:N1
225 - 0 C6: N1
O0C8:N1 A
225 -
Control
0
225 -
225 - A
225 -0
225 A A
225 - {3 
O
225 -
A
225 -
225 -
225 I I I I I I I I
0 20 40 60 80 100 120 140 160
Days
Figure 13. Tilapia growth in ponds fertilized with only chicken litter
(C11:N1, control), or chicken litter with supplemental urea to
maintain C:N ratios at 8:1, 6:1, or 4:1.
TABLE 16. WEEKLY INPUTS OF CARBON AND NITROGEN AS CHICKEN LITTER, AND NITROGEN AS
UREA TO MAINTAIN C:N RATIOS OF 8:1, 6:1 OR 4:1 COMPARED WITH 11:1 FOR ONLY CHICKEN LITTER
(CONTROL). MEAN (? S.E.) GROSS YIELD AND INDIVIDUAL WEIGHT OF OREOCHROMIS NILOTICUS
STOCKED IN 0.1-HA EARTHEN PONDS AT 20,000 MALES/HA FOR 154-DAYS GROW-OUT PERIOD
Chicken litter Supplemental Total N Gross fish Mean weight
Treatment C N N yield
kg/ha kg/ha kg/ha kg/ha g/fish
C11:N1 (Control) 187.8 17.1 0.0 17.1 2,825 97.1 154 ? 10.1
C8:N1 187.8 17.1 6.3 23.3 2,764 192.9 150? 10.1
C6:N1 187.8 17.1 14.1 31.1 3,685 94.1 222 ? 1.49
C4:N1 187.8 17.1 29.9 46.9 2,709 508.8 166 ? 19.5
thick surface scum
formed, and eventually covered half the pond. Fish growth
ceased after 2 to 3 months, although primary productivity
was high and early morning dissolved oxygen concentra-
tions were not lower than other ponds. Stress from a combi-
nation of blue-green algae toxicity and relatively high early
morning total ammonia (1.02 mg/L) and pH (8.9) were the
presumed causative factors of the fish mortality. Pond B7
(treatment C4:1N) also developed a heavy blue-green sur-
face scum, and fish yield (2,191 kg/ha) in this pond was
relatively low compared to 3,186 kg/ha in the remaining
replicate pond, where blue-green algae were not apparent. A
blue-green algae scum developed to a much lesser extent on
pond B5 (treatment C6:N1), but was not encountered on
other ponds in this treatment. These were the first persistent
blooms of blue-green algae recorded at El Carao during the
PD/A CRSP. High nitrogen input apparently promoted blue-
green algae growth.
STUDY B7. SUBSTITUTION OF INORGANIC
NITROGEN AND PHOSPHORUS FOR CHICKEN
LITTER IN PRODUCTION OF TILAPIA
Nitrogen supplementation of organically fertil-
ized fish ponds can increase primary productivity and
fish yields. Results of Study B6, where the C:N ratio of
organic fertilizer was manipulated, suggested that weekly
total N input near 25 kg/ha would increase fish produc-
tion without wasting nitrogen. The objective of this study
was to ascertain if organic inputs could be decreased
through substitution with inorganic N and P.
The study was repeated during cool and warm
seasons of the year. The methodology was equal during
both seasons except where indicated. Ponds were ran-
domly assigned to treatments, each replicated three times.
Chicken litter was applied weekly to ponds at 750 (CL750),
500 (CL500), 250 (CL250), or 0 (CLO) kg DM/ha during
the cool season, and 500, 250, or 0 kg DM/ha during the
warm season. Urea was applied to maintain weekly total
17
a and primary pro-
ductivity, but cor-
relations were not
significant.
Blue-green
algae (Anacystis
sp.) became domi-
nant in pond B10
within six weeks
of stocking. A
TABLE 17. MEAN WEEKLY INPUTS OF UREA AND DIAMMONIUM
PHOSPHATE (DAP) FOR DIFFERENT RATES OF CHICKEN LITTER
APPLICATION (KG DRY MATTER/HA PER WEEK)
Cool season Warm season
Treatment Urea DAP Urea DAP
kg/ha kg/ha kg/ha kg/ha
750 29.7 0 - -
500 37.5 0 33.5 9.5
250 41.3 9.9 37.6 15.4
0 40.6 31.0 39.5 26.8
N inputs at about 25 kg/ha. Diammonium phosphate (DAP)
was added as needed to maintain N:P ratios of at least 4:1.
Manure was frequently analyzed for nitrogen and phospho-
rus. Fifty-five and 100 percent of the N and P, respectively,
in chicken litter was assumed available for plankton use
during the cool season. During the warm season, 50 percent
of N and P in manure was assumed to be available for
phytoplankton. Mean weekly inputs of inorganic fertilizer
during cool and warm seasons are summarized in Table 17.
Nitrogen and phosphorus constituted 2.4 and 1.7 percent,
respectively, of chicken litter dry matter during the cool
season, and 2.4 and 1.3 percent during the warm season.
Ponds were stocked with tilapia fingerlings at
20,000/ha. During the warm season all fish were normal
colored Nile tilapia, but during the cool season, half of the
tilapia were normal colored O. niloticus and half were red
tilapia that had been crossed repeatedly with O. niloticus
(Study D4). Guapote tigre (Cichlasoma managuense) fin-
gerlings were stocked at 500/ha to predate on tilapia repro-
duction. Cool- and warm-season studies began on 5 Septem-
ber 1991 and 12 March 1992, respectively; each study lasted
152 days.
COOL SEASON
(SEPTEMBER 1991 - FEBRUARY 1992)
Mean gross fish yields and individual weights were
not significantly different among treatments (Table 18).
Mean individual weights of normal colored and red tilapia
were similar, but mean gross yield of normal colored tilapia
(1,353 kg/ha) was 75 percent greater than yield of red tilapia
TABLE 18. SUMMARY OF FISH PRODUCTION DATA FROM 0.1-HA EARTHEN PONDS IN HONDURAS. PONDS WERE FERTILIZED
WEEKLY WITH CHICKEN LITTER AT 0, 250, 500 OR 750 KG/HA. UREA AND DIAMMONIUM PHOSPHATE WERE APPLIED WEEKLY TO
MAINTAIN TOTAL N INPUT OF 25 KG/HA AND AN N:P RATIO OF 4:1. DURATION OF EACH EXPERIMENT WAS 152 D
Nile tilapia Red tilapia Guapote tigre* Tilapia Gross fish
Treatment Yield Weight Survival Yield Weight 
Survival Yield Weight Survival offspring 
yield
kg/ha g/fish Pct. kg/ha g/fish Pct. kg/ha g/fish Pct. kg/ha kg/ha
Cool season experiment
CL 0 1,057 a 118 a 87 a 783 a 127 a 63 a 5 31 29 19 1,865 a
CL 250 1,507 a 172 a 88 a 838 a 172 a 50 a 5 34 29 1 2,350 a
CL 500 1,567 a 166 a 94 a 859 a 174 a 49 a 8 30 40 2 2,435 a
CL 750 1,312 a 140 a 94 a 609 a 148 a 41 a 5 35 28 25 1,950 a
Warm season experiment
CL 0 2,045 a 118 b 85 a -- -- -- 5 21 49 29 2,079 a
CL 250 2,317 a 162 ab 71 a -- -- -- 6 32 36 16 2,339 a
CL 500 3,560 a 231 a 77 a -- -- -- 4 58 17 20 3,584 a
* Chichlasoma managuense.
ab Means in columns within experiments followed by the same letter are not significantly different (P>0.05).
TABLE 19. MEANS OF WATER QUALITY VARIABLES IN 0.1-HA EARTHEN PONDS IN HONDURAS. PONDS WERE FERTILIZED
WEEKLY WITH CHICKEN LITTER AT 0, 250, 500 OR 750 KG DRY MATTER/HA. UREA AND DIAMMONIUM PHOSPHATE WERE APPLIED
WEEKLY TO MAINTAIN TOTAL N INPUT OF 25 KG/HA AND AN N:P RATIO OF 4:1. DURATION OF EACH EXPERIMENT WAS 152 D.
Total Organic Total Total Filterable Seechi disk Net primary
Treatment pH alkalinity nitrogen ammonia phosphorus orthophosphate visibility Chlorophyll a production
mg/L CaCO
3  
mg/L mg NH
3
-N/L mg/L mg PO
4
-P/L cm mg/m 
3  
mg 0
2
/L per d
Cool season experiment
CL 0 9.58 a 67.5 c 3.44 c 0.403 b 2.64 b 1.69 a 13.6 a 495 a 9.6 b
CL 250 9.64 a 89.5 b 3.95 b 0.311 b 2.67 b 1.38 a 12.2 a 703 a 11.3 a
CL 500 9.50 a 99.2 b 4.12 b 0.199 a 2.85 b 1.45 a 11.4 a 812 a 12.9 a
CL 750 9.45 a 126.1 a 4.42 a 0.153 a 3.70 a 1.80 a 10.6 a 888 a 12.6 a
Warm season experiment
CL 0 9.85 a 80.7 a 3.14 c 0.323 a 1.99 b 1.16 a 19.5 b 333 b 8.1 b
CL 250 9.45 a 92.0 a 3.59 b 0.280 a 2.69 b 1.37 a 13.5 a 613 a 11.4 a
CL 500 9.61 a 103.6 a 3.90 a 0.169 a 2.77 b 1.30 a 12.2 a 705 a 12.3 a
18
(772 kg/ha). Yield differences between the two types of
tilapia were related to survival. Mean survival of normal
colored tilapia (90.9 percent) was 79 percent greater than
that of red tilapia (50.8 percent). Red tilapia suffered a
greater rate of predation by ospreys as witnessed by field
workers. Also, our experience is that the red tilapia does not
survive handling and environmental stress as well as the
normal colored tilapia (see Study D4).
Covariant analyses of fish growth indicated sig-
nificant differences among treatments. Growth was curvi-
linear for both normal colored and red tilapia in all treat-
ments except CL250, where it was linear. Growth stopped
after the 4th month in the CLO treatment.
Net primary production, chlorophyll a , total-P,
organic-N, total alkalinity, and total hardness significantly
decreased with decreasing litter input; total ammonia and
Secchi disk visibility significantly increased with decreas-
ing litter input (Table 19). There were no significant differ-
ences among treatments for pH or filterable orthophosphate
(Table 19).
WARM SEASON
(MARCH - AUGUST 1992)
Mean individual weight of tilapia in CL500 was
significantly greater than weight of fish in CL250 and CLO
(Table 18). Gross fish yield at the highest level of organic
fertilization was 53 and 72 percent greater than at the
intermediate and lowest levels, respectively. However, dif-
ferences were not statistically significant, with high coeffi-
cients of variation at the intermediate (42 percent) and
lowest (48 percent) organic fertilization levels. Mean yield
of the CL250 treatment was lowered by 55 percent mortality
in one replicate.
Analysis of covariance indicated that the slopes of
mean fish growth curves were significantly different among
the three treatments (P < 0.0001). Fish growth in ponds
receiving no chicken litter practically stopped after three
months. Growth of fish in the two manured treatments was
linear, although growth in the CL 250 treatment appeared to
slow during the last six weeks.
Net primary production, chlorophyll a , total-P,
organic-N, total alkalinity, and total hardness significantly
decreased with decreasing litter input; total ammonia and
Secchi disk visibility significantly increased with decreas-
ing litter input (Table 19). There were no significant differ-
ences among treatments for pH or filterable orthophosphate.
Except for CLO, fish yields were generally lower
during the cold season relative to the hot season. Poorer
yields in the cold season can be attributed to lower survival
and growth. Fish yields for the non-manured treatment were
similar during both seasons; cessation of fish growth before
harvest indicated that carrying capacity had been reached.
Lower temperatures during the cold season delayed achieve-
ment of carrying capacity until the 4th month, whereas it was
reached in three months during the hot season.
Primary productivity was similar during both sea-
sons for each level of CL input. During each season, primary
productivity decreased with decreasing CL input. The re-
duction in productivity was not related to a lack of N and P,
however. Total ammonia nitrogen actually increased with
decreased CL input because of lack of absorption by phy-
toplankton. Filterable orthophosphate was not different
among treatments; furthermore, orthophosphate levels were
high in all ponds and presumably would not limit productiv-
ity. Chicken litter may have provided a micronutrient that
was limiting phytoplankton growth. A more probable expla-
nation was that decomposition of chicken litter made more
CO
2 
available for phytoplankton use. Correlation of alkalin-
ity and hardness levels with increasing CL input supports
this supposition. Carbon dioxide increases the dissolution of
calcite and dolomite, thereby forming bicarbonate (alkalin-
ity) and Ca
2 + 
and Mg
2 + 
(hardness). Higher CL inputs re-
sulted in more CO
2 
evolution, which increased total alkalin-
ity and total hardness. Phytoplankters would have had greater
CO
2 
available to them directly or indirectly through the
bicarbonate-carbonate system. Availability of CO
2 
through
the alkalinity system decreases drastically, however, at the
high pH values measured during these trials, thereby accen-
tuating low CO
2 
problems in non-organically fertilized
ponds.
Primary productivity was clearly benefited by or-
ganic fertilization during both seasons. It appears from the
cold season study that weekly CL inputs beyond 500 kg/ha
were not correlated with higher increases of primary produc-
tivity. Fish yield was not benefited by higher primary pro-
ductivity during the cold season, because colder tempera-
tures and reduced fish densities would not allow fish biom-
ass to take advantage of higher nutrient availability. How-
ever, fish clearly benefited from higher natural productivity
during the hot season.
Substitution of inorganic N for organic inputs has
allowed for weekly chicken litter fertilizer rates to be re-
duced from 750 to 500 kg/ha without a decrease in fish
production. Less organic input implies less biological oxy-
gen demand, which in turn may decrease community respi-
ration rates that have resulted in nocturnal hypoxia. Results
of Study E2 demonstrated that chronic hypoxia can decrease
tilapia yields.
19
COMBINED FERTILIZATION AND FEED STUDIES
Many fish farmers use commercially -formulated
rations in order to obtain faster growth and greater fish
yields. Fish feed is expensive and may not actually improve
profitability of fish farming, even if greater yields and larger
fish result. Published data indicate that efficiency of nutri-
tionally incomplete feeds can be increased greatly if supple-
mented by natural food organisms stimulated by pond fertili-
zation. We undertook a series of studies to quantify how best
to combine feeds with fertilizers in semi-intensive tilapia
monoculture ponds. Formulated feed, a pelleted marine
shrimp feed, used in this research was procured locally.
Economic evaluation of management systems tested is re-
ported in the economic analysis section of this report.
STUDY CL EFFECT OF STOCKING RATE ON YIELD
OF TILAPIA OFFERED A FORMULATED DIET
Stocking rates for monosex growout of tilapia
traditionally range from 10,000 to 20,000 fish/ha in ponds
managed with neither water exchange nor mechanical aera-
tion. Under these conditions 150- to 400-g fish are harvested
after five to six months of grow-out. In rural areas and in
smaller cities in Honduras fish weighing 150 to 200 g are
easily sold, while larger fish (400 g) are sold more easily in
metropolitan areas. It is well known that stocking rate affects
the average size of fish at harvest, but this relationship is
influenced by the quality and quantity of nutrient input. The
objective of this study was to determine the effect of
stocking rates on monosex production of tilapia offered a
prepared ration.
Nine ponds were used for this completely random-
ized design study where stocking rates of 10,000, 20,000 and
30,000 tilapia/ha were tested. Tilapia fingerlings with aver-
age weight of 17 g were stocked into ponds on 17 July 1987.
Ponds were fertilized once with chicken litter (1,000 kg DM1
ha) upon initiation of the experiment. Fish were fed a 23
percent protein pelleted ration six days per week. Feeding
10,000 15,00 20,000 25,00 30,000
Stocking rate (fish/ha)
Figure 14. Relationship between gross tilapia yield and stocking
rate in 0.1-ha earthen ponds. Experiment lasted 150 days and
fish were fed a 23 percent protein formulated diet.
Average individual weight
(g/fish)Y 
= 287.367 - 0.0041x
250 
r
2
=- 0.928
230
210
Figure 15. Relationship between mean individual fish weight
at harvest and stocking rate. Nile tilapia were reared for 150
days in 0.1-ha earthen ponds and fed a formulated ration (23
percent protein).
percent during the first month, 5 percent during the second
month, and 3 percent during months three to five. The
feeding rate was adjusted monthly based upon seine samples
and assuming 100 percent survival. All ponds were har-
vested 150 days after stocking. The feed conversation ratio
(FCR) was calculated from the total amount of feed offered
fish divided by net fish production.
TABLE 20. SUMMARY (MEAN + S.E.) OF HARVEST DATA AND FEED CONVERSION RATIO
(FCR) FOR PRODUCTION OF MALE OREOCHROMIS NILOTICUS STOCKED INTO 0.1-HA EARTHEN
PONDS AT THREE RATES. FISH WERE FED A 23 PERCENT PROTEIN PELLETED RATION
Stocking rate Final size Survival Gross yield Net yield FCR
fish/ha gifish Pct. kg/ha per 150 days
10,000 249 ?0.7 a 97.l?1.8 a 2A410 ?47 a 2,245 ?46 a 2.0 ?0.11la
20,000 200 ?9.9 b 92.5 ?0.8 a 3,708 ?188 b 3,379 ?186 b 2.7 ?0.01lc
30,000 166 ? 3.9 c 96.4 ? 1.9 a 4,817 ? 10 c 4,323 ? 16 c 2.4 ? 0.05 b
abc Column means followed by the same letter were not significantly different (P>0.05).
Mean gross fish produc-
tion increased significantly as stock-
ing rate increased (Figure 14), vary-
ing from 2,4 10to 4,8 17kg/ha (Table
20). Tilapia survival was similar
among treatments, averaging 95
percent. Mean individual fish size
ranged from 166 to 249 g at harvest,
and decreased linearly as the stock-
ing rate increased (Figure 15). Fish
production was proportional to
20
Gross tilapia yield
(kg/ha per 150 d)
000
y 0.12x - 1238.33
r
2
-0.973
4,500
4,000
3,500
3,0001
2,500
190
170
150
10,0000 00 20,000 25,000 3
Stocking rate (fish/ha)
~n nnn
rt30)000
socki:-gc rate because, the tilapia biomass ed no reach 
e
critical standip- crop in any trea tme:,. Mea CR va ried
Irom 1.9 w 2.74. The nigh FCR observed 
17 te 2,,000
tila-iaf-a treatmrent may hav e been infliuncec by 
lower
observe fis su- 5Lval
Since no preferentialprc has bee---- es, abisid 
"of
larger fish tOoc de, production a- the 10,000 
tiiapia/'ia ra~te
was lotjustifieel. M~ean isszee neihsa& 
s
near to the apparent inimumo siz Caccepted 
bb con'suo ers
and culture during cooler times d' the year 
may insuh i
smallerfish, making commercializae tonnroorei 
d'ffiu . High,
yields of readily mrarketable fish were obtained 
at a stocking
rate oC 20,000 tilapiafna.
STUDY C2. ThRGDUCINI c,- j-, i APE C\UJ3
C-rVJBINAT CNS OIF C HC7c,\ AND~ FEED~
Previous exper'ence indica ted the tilapia g-vt
in organically fertiiized ponds dec-cased win -htre 
b-ecause
available natural food no longer sustained fast grov./th 
Fedc
I za-,on efficiency might beimproveo ife fertiliizer cole
be usedi to sO imutae natural pond productiviiy durirg h
early part of lie cultre -period, fol owed by a more expen-
sive, formeleteC eed as 'he cultir period progressed. 
Tile
ojective of th- study was to ies-L te effects o' sev 
era-
enombinations of chicker. liter ad -fedh bi the
1 
dcdo
tilapia.
Nine no -ids were used forthis corleiclyrino Th o
izeel design study where the production syster. s esed were.
fee, only (FO); layer chicken litter only (1,00) Ig 
DIMTha
per wek) fur the first 60 days, followed by fee o nly (3
percent oiltilapie iomass) (CL/7); layer chicken i 'er (500
kg DM/Iha per vi nek) are' feed (1.5 percent of the tilaoia
biornss) throogo u 1,he -ura, i on o the ex-pe.-emjCe -).
Tiao: igerl ags (mear weight CC l9 g) were stocked in- o
pondPs at 20,000 f'shb/a o.i 20 Janu ary 989. Fertilization
'JA 2 fL (0 7 N A~ L 3-i~ I , V--7-,A- IF C,
(FCR) FOR -nROU-3 OF MAL Vin C'3' 'ora!(0
VeatmenL Fifla size Surviit Crose Nti
,fish act.
262 ?19.l a 86.9 ?2.6 a
251 ? 23 a 82.3 +2.3a
5,305 + 351a 4,96 15S
4,351?+22)Ca 3,984 +Z2 8
--asei ( e -ce ' pc. Cc y ee
-,,/ars c ' 'D G7, hD 'C Ce / S -- 'ie s or
,h 1'c 7 ' rae a . peK. o ' 'h K C KS i'o
ea s~ ,, 0 dr s
0  
i. 1m Csse 1
At na-Cv st't 'C :Lr, n ''Cv '1 ,s ' s C C
52~ f i'V " .JC .5'sr ' C
stat -- ! '--h V s mc D5i .cs I2 vve5grC
eng - 4,1 5 . 2 C1 t- Ch
i + C(0.
C+ '
F 0: feed on ly CL +-F: Iayarch ick en liter(5 00o ,f- a, peawe ek) and 
1,C --6 b i Oc T3,Ig
out the duration o!he exiperiment; CF; layer chicken liter only (1,000O 
kl- DI' per
week) for the first 60 days, followed by feed only.
ab Column mears followed by the sane letter wer no', signiicantly di 
f arn >J.,
- ,'n 2 - -'
te.,Fr 2-7 t2
.110vin C ----1 -'J /a sC' K
i 1. " ficant r- 1la
C-'
C'C
(T - , -- i
e 's C .i>
~ a:
FO
C7
FCR
resulted from primary production, whereas primary produc-
tion and feed contributed to fish growth in the other two
treatments. After 60 days, mean individual weight in the FO
treatment (94 g/fish) was significantly greater than the 74 g/
fish in the CL/F treatment, which indicated that available
natural food in CL/F ponds was insufficient to maintain
rapid growth. Although gross primary productivity was
significantly greater for the CL/F treatment during the first
60 days (Table 22), the addition of feed thereafter was
necessary to maintain fast growth. Mean tilapia standing
crop on day 61 in CL/F ponds was estimated at 1,496 kg/ha.
Initiation of supplemental feeding in the CL/F treatment
significantly increased tilapia growth from a mean of 1.0 g
per day on day 61 to 2.0 g per day on day 91. At harvest, no
significant difference in mean fish size among treatments
was observed and average final size for fish from all treat-
ments was 265 g/fish.
Feed conversion ratio (FCR) varied significantly
among treatments (Table 21). FCR for the feed only treat-
ment (1.8) was greater than the 1.0 FCR for the chicken litter
plus feed treatment. The FCR of 1.0 in the latter treatment
was attributed to the contribution of natural food, stimulated
by the organic fertilization. An intermediate FCR was ob-
served for the chicken litter followed by feed treatment,
indicating feed was not necessary during the first two months
TABLE 22. MEAN (? S.E.) PRIMARY PRODUCTIVITY AND COMMUNITY
RESPIRATION (G 0
2
/M
3 
PER DAY) FOR SPECIFIED TIME
INTERVALS IN 0.1-HA EARTHEN PONDS STOCKED WITH MALE
OREOCHROMIS NILOTICUS (20,000/HA) AND RECEIVING FEED ONLY
AND COMBINATIONS OF CHICKEN LITTER AND FEED
Primary productivity 
Community
Treatment' Gross Net respiration
DAYS 1 to 30
FO 7.74 ? 0.34 b 3.19 ? 0.37 b 7.74 ? 0.34 b
CL + F 9.47 ? 0.42 b 4.37 ? 0.45 ab 9.47 ? 0.42 ab
CL/F 13.39 ? 0.42 a 6.46 ? 0.45 a 13.39 ? 0.42 a
DAYS 1 to 60
FO 
6.98 ? 0.47 
c 2.94 ? 
0.25 c 8.07 
? 0.82 c
CL +F 9.91 ? 0.58 b 4.39 ? 0.31 b 11.04 1.00 b
CL/F 13.72 ? 0.58 a 6.61 ? 0.31 a 14.21 1.00 a
DAYS 61 to 151
FO 11.69 ? 1.06 b 6.33 ? 0.74 b 10.72 1.25 b
CL + F 13.77 ? 1.30 b 7.47 ? 0.91 a 12.61 1.53 ab
CL/F 16.80 ? 1.30 a 9.45 ? 0.91 a 14.76 ? 1.53 a
151-d GROW-OUT PERIOD
FO 10.12 ? 1.42 b 5.20 ? 1.07 b 9.84 1.20 c
CL + F 12.49 ? 1.74 b 6.44 ? 1.31 b 12.09 1.47 b
CL/F 15.77 ? 1.74 a 8.50 ? 1.31 a 14.57 ? 1.47 a
1 FO:feed only; CL + F: layer chicken litter (500 kg DM/ha per week)
and feed, both throughout the duration of the experiment; CL/F: layer
chicken litter only (1,000 kg DM ha per week) for the first 60 days,
followed by feed only.
abc Means followed by the same letter were not significantly differ-
ent (P>0.05). Vertical comparisons within time period only.
Oreochromis niloticus fingerlings (18 g) were
stocked into ponds at 10,000/ha. Twelve ponds were ran-
domly assigned to four treatments. Treatments were: chicken
litter only, or chicken litter plus feed (23 percent protein) at
0.5, 1, or 2 percent of fish biomass. Chicken litter was
applied to ponds weekly at 500 kg DM/ha. Feed was offered
to fish six days per week, and quantities were adjusted
weekly based on projectedgrowth between monthly samples
of fish weight. The experiment was initiated on 11 August
1988 and was terminated 132 days later.
At harvest, mean individual weights were 132,136,
162, and 170 g/fish for 0, 0.5, 1, and 2 percent feed treat-
ments, respectively. Mean gross yields were 1,234, 1,261,
1,473, and 1,604 kg/ha per 132 days, respectively. Mean
individual weight and gross yield for the 1 and 2 percent feed
treatments were significantly greater than for the other two
treatments. Fish production was low compared to other
years at El Carao, especially considering that fish were
offered a commercial ration in addition to fertilization. This
study was carried out during the cold season when water
temperatures frequently fell below optimal growth levels.
Water temperatures as low as 17.4 'C were recorded. Grow-
out during the hot season would have provided better infor-
mation regarding potential benefits of supplemental feed-
ing.
22
of growout. Substituting feed with intensive, organic
fertilization during the first two months of growout
resulted in a 27 percent reduction in total feed require-
ment compared to the feed only treatment. Further feed
savings were achieved with the chicken litter plus feed
treatment where the mean total feed utilization was
3,764 kg/ha, or 42 percent of that for the feed only
treatment. Results of economic analysis (economic
analysis section) indicated that net returns were higher
where chicken litter was substituted for feed. It should be
noted that the FCR for the feed only treatment was less
than the FCR reported for the same stocking rate in Study
C1, implying that the initial feeding rates, of 7 percent
and 5 percent of the tilapia biomass, were high.
STUDY C3. SUPPLEMENTAL FEEDING OF
TILAPIA AT VARIOUS RATES
Study C2 demonstrated that a combination of or-
ganic fertilization and supplemental feeding resulted in
similar fish yields and better feed conversion than where
feed only was used. The objective of this study was to
determine a practical rate of supplemental feeding to
increase profitability.
STUDY C4. DELAYS IN INITIATION OF
SUPPLEMENTAL FEEDING
In Study C2, initiation of supplemental feeding 60
days after stocking yielded 12 percent less fish but required
27 percent less feed than the treatment where feed was
offered from the beginning of the trial. These results sug-
gested that supplemental feeding might be initiated even
later, especially at lower stocking rates, without reducing
fish yield. The objective of this experiment was to determine
the effect of substituting chicken litter for feed for progres-
sively longer periods on tilapia yield.
Four treatments were assigned to 12 ponds. Treat-
ments tested were: (1) weekly applications of chicken litter
at 1,000 kg DM/ha; (2) chicken
litter (1,000 kg ha per week) for
the first month, followed by feed-
ing; (3) chicken litter (1,000 kg
DM ha per week) for two months,
followed by feeding; and (4)
chicken litter (1,000 kg DM ha per
week) for three months, followed
by feeding. Feed (25 percent pro-
tein) was offered six days per week
at 3 percent of fish biomass per
ciency could be increased significantly by substituting or-
ganic fertilization for feeding during the early part of the
grow-out cycle. Tilapia yields in this study were lower than
those attained in earlier research where chicken litter was
substituted for feed because a lower stocking rate was used.
Tilapia stocking rate should be at least two fish/m
2 
when
feed is used (economic analysis section).
OTHER SPECIES STUDIES
Monosex culture of tilapia has been practiced in
Honduras since the early 1980s. While the goal of monosex
production has been to eliminate the production of tilapia
offspring during the grow-out cycle, a few female tilapia
were stocked inadvertently and resulted in some reproduc-
tion. A native predator fish was se-
lected to be co-stocked with tilapia to
prey upon tilapia offspring. Stocking
of more than one species with non-
overlapping food habits results in
greater fish yields. A series of studies
was conducted to investigate
polyculture systems of interest in
Honduras that were based on native
and exotic species.
TABLE 24. SUMMARY OF TILAPIA REPRODUCTION AND GUAPOTE TIGRE YIELD RESULTS
WHEN GUAPOTE WERE STOCKED AT FIVE PERCENT OF STOCKED TILAPIA
TO PREY UPON TILAPIA OFFSPRING PRODUCED
Guapote
Tilapia Guapote Tilapia
stocking rate stocking rate reproduction Mean weight Gross yield Survival
No./ha No./ha kg/ha g/fish kg/ha per 150 d Pct.
2,500 .................. 125 232 a 172 a 16.7 a 65.0 a
10,000 .................. 500 11 b 117 b 36.1 b 61.5 a
20,000 .................. 1,000 2 b 87 c 53.2 c 61.3 a
abc Means within columns followed by different letter are significantly different (P < 0.01).
day. Ponds were stocked with tila-
pia (average weight 28 g) at 10,000/ha and guapote tigre
(Cichlasoma managuense) at 450 fish/ha to control tilapia
reproduction. Ponds were stocked on 27 July 1989 and
harvested 147 days later.
Gross tilapia yield and mean individual weight
were significantly less in ponds fertilized with only chicken
litter (Table 23). Initiating feeding after one month, how-
ever, provided no significant advantage over waiting three
months to begin feeding. These results confirmed those of
earlier studies at El Carao, which demonstrated that tilapia
yields could be increased with feeds, but that feeding effi-
STUDY D1. RELATIONSHIP BETWEEN GUAPOTE
TIGRE STOCKING RATE AND
TILAPIA REPRODUCTION
Sex reversal and manual separation of male fish
from mixed-sex populations usually are not 100 percent
effective in excluding females from the grow-out pond.
Guapote tigre (Cichlasoma managuense), a native cichlid
piscivore, were included regularly in grow-out ponds after
1987 to control tilapia reproduction. Guapote initially were
stocked at rates of one per 10 to 20 tilapia, depending on
23
TABLE 23. MEAN (? S.E.) GROSS YIELD, FINAL INDIVIDUAL WEIGHT AND SURVIVAL
OF TILAPIA STOCKED AT 10,000/HA IN 0.1-HA EARTHEN PONDS THAT RECEIVED
COMBINATIONS OF CHICKEN LITTER AND FEED APPLICATIONS
Treatment Gross yield Individual weight Survival
kg/ha per 147 days g/fish Pct.
Chicken litter only 1,779 + 80 a 206 ? 9.1 a 86 ? 0.1 a
Chicken litter 1 month, then feed 2,349 ? 142 b 276 ? 16.2 b 85 ? 3.0 a
Chicken litter 2 months, then feed 2,196 + 101 b 256 ? 8.6 b 85 ? 1.2 a
Chicken litter 3 months, then feed 2,223 + 170 b 258 ? 16.7 b 86 ? 2.0 a
ab Means within column followed by the same letter were not significantly different
(P<0.10)
guapote availability. The objective of this study was to
determine if optimum stocking rate of a predator was a
function of tilapia stocking rate.
Twelve ponds were assigned randomly to three
treatments. Treatments tested were tilapia stocking rates of
2,500, 10,000, and 20,000/ha. Tilapia fingerlings (29 g)
were stocked in ponds on 3 February 1989. Two replicate
ponds per treatment were stocked with sex-reversed males,
and the remaining two ponds were stocked with males that
had been manually selected from a mixed-sex population.
Guapote fingerlings (22 g) were stocked in all ponds at one
fish per 20 tilapia stocked. Ponds were fertilized weekly with
CL at 750 kg DM/ha. Ponds were harvested after 150 days.
Tilapia production data are reported in Study E3.
Guapote controlled tilapia reproduction more ef-
fectively at the moderate and high tilapia stocking rates than
at the low stocking rate (Table 24). Survival was not differ-
ent among treatments. Mean individual guapote weight
increased with decreasing tilapia stocking rate, while guapote
yield increased with increasing tilapia stocking rate. We
concluded that insufficient guapote to control reproduction
were stocked at the low tilapia stocking rate, while more
guapote than necessary were stocked in the high tilapia
stocking rate treatment.
Adequate numbers of guapote are needed to patrol
a given pond area regardless of tilapia stocking rate. Insuf-
ficient numbers of guapote will result in poor reproduction
control, while excessive numbers of guapote stocked will
result in small guapote at harvest. Large numbers of similar-
sized guapote fingerlings are difficult to produce in ponds
(Study A4) because they are predacious. Therefore, the
minimum number of guapote necessary to control reproduc-
TABLE 25. SUMMARY OF HARVEST DATA (MEAN + S.E.) FROM 0
EARTHEN PONDS STOCKED WITH NILE TILAPIA AND TAMBAQUI (COL
MACROPOMUM) THAT RECEIVED WEEKLY APPLICATIONS OF ORGANIC F
AND WHERE FISH WERE OR WERE NOT OFFERED A SUPPLEMENTA
Treatment
Organic fertilization Organic fert
Variable only plus supplem
Tilapia
Gross yield, kg/ha per 126 d 1,355 ? 102 1,921 ?
Individual weight, g/fish 167 ? 10 232 ?
Survival, pct. 81 ? 2.3 83 ?
Tambaqui
Gross yield, kg/ha per 126 d 76 ? 22 350 ?
Individual weight, g/fish 86 ? 27 447 ?
Survival, pct. 60 ? 2.6 52 ?
tion are desired for stocking with tilapia. The stocking rate
of 500 guapote/ha resulted in good control of tilapia repro-
duction. Subsequent tests at this same guapote stocking rate
have yielded consistently good results, even when indi-
vidual size of guapote fingerlings stocked was reduced to
one to five g.
STUDY D2. INFLUENCE OF FEED AND ORGANIC
FERTILIZATION ON POLYCULTURE OF
TAMBAQUI AND TILAPIA
In Honduras there was interest in determining the
growth response of tambaquf (Colossoma macropomum)
when co-stocked with tilapia in organically fertilized ponds.
There was evidence that tambaquf did not grow well in the
absence of supplemental feed. The objective of this experi-
ment was to quantify growth of tambaquf in polyculture with
tilapia in ponds receiving only organic fertilization or a
combination of organic fertilization and supplemental feed.
Six ponds were stocked on 13 February 1990 with
tilapia (13-g average weight) at 10,000/ha, tambaquf
(Colossoma macropomum; 40-g average weight) at 1,500/
ha, and guapote tigre (Cichlasoma managuense) at 250/ha to
control tilapia reproduction. Three ponds were fertilized
weekly with chicken litter at 1,000 kg DM/ha for the first six
weeks, and then at 750 kg TS/ha for the remainder of the
study. The remaining three ponds were fertilized weekly
with chicken litter at 500 kg DM/ha, and fish were offered a
pelleted ration (25 percent protein) at about 1.5 percent of
tilapia biomass per day, six days per week. Ponds were
harvested on 19 June 1990, 126 days after stocking.
Mean average weights of tilapia and
tambaquf, and mean gross fish yields were
.1-HA significantly greater in fed ponds (Table 25).
OSSOMA 
Tambaquf 
were 422 percent 
larger where 
supple-
ERTILIZER
L FEED mental feed was used compared to 
organic
fertilization alone, whereas tilapia were only 39
percent larger in the supplemental feed treat-
ilization ment. Tambaquf grew little and did not reach a
ental feed marketable size in 
ponds that received only
organic fertilization. The optimal stocking ratio
40* is unknown. The practical implication of this
16* study is that fish farmers should not stock
4.4,
tambaquf unless a formulated ration is used.
70*
71*
2.3
24
*Treatment means were significantly different (P<0.05). Horizontal
sons only.
I compari
\ I
STUDY D3. GROWTH OF CICHLASOMA
MACULICUADA CO-STOCKED WITH
TILAPIA AND GUAPOTE TIGRE
STUDY D4. COMPARATIVE GROWTH OF
COMMUNALLY STOCKED RED
AND WILD-TYPE TILAPIA
Cichlasoma maculicauda is a native cichlid found
at sizes greater than 300 g in lakes and rivers in Honduras.
Fingerling Cichlasoma maculicauda (five to 25 g) were
obtained from a fingerling production facility in Olancho
(Eastern Honduras). This fish has a small mouth and filiform
teeth, and was reported to graze on the pond bottom, for
which it appeared to be a suitable fish for pond culture. The
fingerlings were stocked in a fertilized pond and received a
daily ration of formulated feed. After 9 to 12 months in the
pond reproduction was observed. Fingerlings produced in
the pond were used in a subsequent grow-out trial.
Cichlasoma maculicauda (12-g average weight)
were stocked at 290/ha with tilapia and guapote in a study to
determine the effect of minimal oxygen concentration (Study
TABLE 26. MEAN (+ S.D.) YIELD AND SURVIVAL OF CICHLASOMA
MACULICUADA IN 0.1-HA EARTHEN PONDS THAT RECEIVED NO
MECHANICAL AERATION OR AERATION THAT BEGAN WHEN
DISSOLVED OXYGEN CONCENTRATIONS WERE 10 PERCENT
OR 30 PERCENT OF SATURATION
Treatment Mean weight Yield
(g/fish) Pct. survival (kg/ha per 148 d)
Control 53 ? 12.9 47 38 8 7
10% saturation 80 ? 5.6 64 5.3 15 0.3
30% saturation 87 ? 4.2 79 15.0 20 3
Contrast
Control vs. aeration HS* NS S
10% vs. 30% NS NS NS
*NS - not significant (P>0.05); S - significant (P<0.05); HS - highly
significant (P < 0.01).
E4). Ponds were organically fertilized for the first two
months with chicken litter (1,000 kg DM/ha per week).
Thereafter, fertilization was stopped and fish were offered a
20 percent-protein pelleted feed. Ponds were harvested after
148 days.
Growth of Cichlasoma maculicauda was slow in
all treatments, despite aeration and a combination of natural
foods and formulated diet (Table 26). Mean individual
weight at harvest did not exceed 83 g, despite the low fish
stocking rate. Low dissolved oxygen concentrations ap-
peared to inhibit Cichlasoma maculicauda growth and sur-
vival. Poor performance of Cichlasoma maculicauda in this
trial indicated their growth is too slow for commercial
aquaculture.
Red tilapia is rapidly becoming popular as a culture
fish in Honduras. El Carao has had two different lines of red
tilapia with greatly different appearances and culture char-
acteristics. The first line, introduced from Mexico, was
presumed to have originated from a commercial producer of
red tilapia in Florida. This fish was pink and bred true, but
because of low fecundity, poor growth and low resistance to
handling, its use was discouraged. In 1989, a commercial
farm in Honduras imported a red tilapia from Geneva,
Alabama; this fish apparently originated with L. L. Behrends,
National Fertilizer Development Center in Muscle Shoals,
Alabama, who back-crossed several generations of red off-
spring with Oreochromis niloticus parents. The red color
originated from the Florida red tilapia. The Alabama red did
not produce 100 percent red off-spring, but its growth and
resistance to handling were superior to the Mexican line.
The objective of this study was to compare growth
characteristics of the Alabama red strain to wild-type O.
niloticus.
Three ponds each were stocked communally
on 8 August 1990 with 500 red and 500 wild-type male
tilapia. Red and wild-type fingerlings averaged 8 and 7
g, respectively. Ponds were fertilized with chicken litter
(1,000 kg DM/ha per week) for the first two months.
Thereafter, fertilization was stopped and fish were fed a
commercial shrimp ration (20 percent protein) 6 days
per week at 3 percent of fish biomass. All ponds received
equal quantities of inputs. Ponds were completely drained
and fish harvested 148 days after stocking.
Mean gross yield of wild-type tilapia (1,133
kg/ha) was significantly greater than that of red tilapia (456
kg/ha). However, mean individual weight of wild-type (252
g) did not differ from that of red tilapia (253 g). Yield of red
tilapia was significantly affected by lower survival, which
averaged 37 and 83 percent, for red and wild-types, respec-
tively. Red tilapia survival was low partly because of preda-
tion by ospreys that visited ponds daily. The ponds farthest
away from human activity suffered the poorest survival.
Lack of resistance to handling, especially at the fingerling
stage, also was suspected of reducing survival. Results of
other studies with red tilapia showed these fish to survive
poorly in growout ponds, even in the absence of bird preda-
tion (see Study B7). These results indicated that growth
potential of red tilapia was similar to that of wild-types, but
consistently low survival would make this red tilapia less
suitable for pond culture. Low ambient temperatures during
the last two months of the cycle probably slowed growth and
contributed to reduced total yields.
25
STUDY E2. IEFFECT OF POND SIZE ON
TILAPIA PRODUCTION
2 '. Bar- Greer assemc r co edrSar~mpler.
MISCELLANEOUS STUDIES
Economic evaluation of management systems tested is re-
ported in the economic analysis section of this report.
STUDY EL. PRODUCTION OF NILE TIILAPA AND
HYISRED 'TYLAPIA IN EARTHEN PONDS
In the early 1980s monosex tilapia culture in H-on-
duras was based on manually-selected Nile tilapia or 
hybrid
tilapia (0. nilotr us x 0. urolepis horniorumq. There 
were
conflicting reports regarding production characteristics of
the two fish, which were theoretically all male. 
The objective
of this research was to compare growth and production 
of
male Nile tilapia to male hybrid tilapia.
Each treatment was randomly assigned to three
ponds. Fish were stocked into ponds at 10,000 fish/ha on 31
August 1983 and harvested after 90 days. Nile and hybrid
fingerlings averaged 7 and i 1g respectively. Ponds were
fertilized on six occasions with 46 kg/ha of 20-20-0 (N-P-K).
Fish were fed a fern alated fish ration (23 percent protein; 3
percent of fish biomass) five days per \x, eek.
Net fish yie' ds were 1,197 and I, 4~1 k-/ha for Nile
and hybrid tilapia, respectively, and did not differ signifi-
cantly between treatm-ents. Likewise, mean individual weight
at harvest were 134 and 131 g, respectively, and did not
differ significantlv. Although the hybrid tilapia initial mean
weight was significantly greater. an analysis of covariance
indicated no significant effect on final individual weight.
Survival av eraged 94.9 percent and 95.6 percent tor Nile and
hybrid tilapia. respectively. While, no differences were
observed in production characteristics between the two fish,
maintenance of the iwo pure-line species required for hybrid
production is more difficult than where no hybrid is pro-
duced. Further, pure-line fingerling production is more
orolitic.
Aquacultural research frequently is carried out in
small (< 0.1 ha) experimental ponds. However, research
ponds generally are smaller than ponds uperated by com-
mercial fish farmers. The purpose of this study was 
to
determine the eftec.L of pond size on fish yield.
Three 0.05-ha and three 0.2 L c ponds were stocked
with fingerling hyhrid tilapia (0. ;iiloticus x 0. urolepis
hornoruma average weight = nine g/fish) at 20,000 fishiha.
Ponds had similar histories of prev ious management. All
ponds were constructed with imported soiL ponds soils were
assumed to he similar. Ponds were fertilized weekly with
chicken litter (38 1 kg DM/ha). Fish were fed corn gluten 
(12
percent protein, 4 percent fat. 16 percent carbohydrate, 5
percent ash and 12 percent moisture) at 7 percent of the fish
hiomass Monday through Friday. Ponds were stocked on 9
Septemher 1985 and harvested 126 days later.
Gross yield averaged 2.502 and 2,736 kg,/ha for the
500- and 2,000-in 2 ponds, respectively. Treatment means
were not significantly different. Mean individual weight 
at
harvest was 150 and 159 g-, respectively. Tilapia offspring
p~roduced during the growout cycle represented 9 percent of
each treatment's gross yield. Survival averaged 76 percent
and 79 percent for the 0.05- and 0.2-ha ponds, respectively,
and did not differ significantly. We concluded that pond size,
within the size range tested, did not affect fish yield in the
two treatments.
STUDY 1E3. TESTING FOR ANABOLIC GROWTH
RESPONSE OF ANDROGEN -TREATED
TILAPIA DURING TREI 
TMENT9
NURSERY AND, GROW-OlUT PHASES
The predominant monosex fingerling production
technology currently employed in Central America is hor-
monal sex reversal, where an androgen, generally 17-(x
methyltestosterone (NIT), is incorporatLed into locally avail-
able commercial rations. MIT treatment for sex reversal of
tilapia may also induce an anabolic growth response. The
objective of this study was to compare the growth of control
and MT-treated tilapia during consecutive androgen-treat-
ment, nursery, and g-row-out phases in earthen ponds under
conditions that likely would be used on commercial, semi-
intensive tilapia farms in Central America.
Recently hatched Oreochrornis niloticus fry were
stocked into hapas suspended in a 0.2-ha earthen pond.
Control and MT-treated treatments each were randomly
assigned to four large (2 x 2.5 x 1 m) and two small (1 x 2 x
1 m) hapas (1.6-mm ace nylon mesh). Water depth in hapas
averaged 0.6 m. Hapas were stocked 
with 4,500 fry/m
2
.
Average total length (? SD) for 500 fry was 9.8 ? 2.3 mm.
Total weight of 500 fry was 4.65 g. Water temperature
during the hormone-treatment period averaged 26.1 'C.
Fry in both treatments were fed a ground, sieved
(560-t mesh) commercial shrimp diet (21.4 percent crude
protein). MT was incorporated into the sex-reversal diet at a
rate of 60 mg/kg feed by dissolving it in 0.5-L denatured
ethanol/kg feed, and thoroughly mixing with the feed. Con-
trol feed was mixed with 0.5-L androgen-free denatured
ethanol/kg feed. Both feeds were oven dried and refrigerated
until fed. Fry were fed seven days per week at a daily rate of
20 percent tilapia biomass during the first week, decreasing
progressively to 10 percent of the biomass during the fourth
week. Hapas were harvested 28 days after stocking on 28
October 1988.
Control or MT-treated fry were stocked into two
0.2-ha earthen ponds each on 28 October 1988 at a rate of
125,000 fry/ha. Fry in both treatments averaged 0.1 g. Prior
to stocking, ponds were fertilized once with 2,000 kg DM/
ha of layer chicken litter. Chicken litter was applied weekly
at 1,000 kg DM/ha during the first month and then at 500 kg
DM/ha thereafter. Fingerlings were fed a pelleted diet (23
percent crude protein) six days per week. Daily feeding rate
was 10 percent of mean fish biomass per treatment during
the first month, decreasing progressively until a maximum
daily feeding rate of 55 kg/ha was achieved. Water tempera-
ture in ponds averaged 24.5 'C. All ponds were harvested by
draining 94 days after stocking.
Growth and yield during growout of male control
and male MT-treated fish at three stocking rates were inves-
tigated using a completely randomized design in 2 x 3
factorial arrangement. Twelve ponds were stocked with
2,500, 10,000, or 20,000 fish/ha on 3 February 1989. Initial
individual weight averaged 29 g for both control and MT-
treated fish. Chicken litter was broadcast over the surface of
all ponds weekly at a rate of 750 kg DM/ha. Guapote tigre,
Cichlasoma managuense (22 g/fish) was stocked into ponds
on 7 February 1989 at a rate equivalent to 5 percent of the
tilapia stocked to control tilapia reproduction. Ponds were
harvested by draining after 150 days. Water temperature in
ponds averaged 26.6 ?C.
Androgen-treatment and nursery phase growth data
were logarithm transformed for linear regression analysis.
Survival data were arcsine transformed for analysis. Data
were analyzed using t-test (treatment and nursery phases),
350
300
Mean weight (g/fish)
400 -
O MT-treated (2,500/ha) 0 Control (2,500/ha)
- = 2,41x + 16.49 y - 2.23x + 20.71
r
2  
0.994 r
2 
= 0.994
- K MT-treated (10,000/ha) Control (10,000/ha)
y = 1.30x + 25.61 y - 1.47x + 19.32
- r
2 
= 0.995 r
2 
= 0.970
-A MT-treated (20,000/ha) A Control (20,000/ha)
y = 0.81x + 27.08 y = 0.77x + 28.45
r
2
0.995 r
2
= 0.992
250
200
150
100
50
I I I I I I I I I I I I I I I I I
100 120 140 1600 20 40 60 80
Days
Figure 16. Growth of control and MT-treated males Oreo-
chromis niloticus stocked at three rates in 0.1-ha earthen ponds
during the 150-day grow-out phase. Ponds were fertilized
weekly with chicken litter (750 kg DM/ha).
Chi-square analysis (nursery phase sex ratio), two-factor
ANOVA where the factors were treatment and stocking rate
(grow-out phase), contrasts (grow-out phase), linear regres-
sion (all phases), and covariance analysis for heterogeneity
of slopes (all phases).
Slopes of control and MT-treated fry growth curves
were not significantly different during androgen treatment.
After the 28-day treatment phase, control and MT-treated
fry both averaged 0.1 g/fry. Final total length did not differ
significantly between treatments and was 18.3 mm and 18.9
mm, respectively.
No significant difference between slopes of growth
curves for control and MT-treated fingerlings was detected
during the nursery phase. No significant differences in
harvest data were detected between treatments. At harvest,
mean individual fingerling weights were 22.5 g and 26.2 g
for control and MT-treated fish populations, respectively.
However, respective mean weights for males were 29.5 and
29.7 g/fish. Females present in both populations were sig-
nificantly smaller than males. Control and MT-treated fe-
males averaged 21.5 and 23.7 g/fish, respectively. Control
27
TABLE 27. PRODUCTION DATA FOR HAND-SELECTED AND SEX REVERSED MALE
OREOCHROMIS NILOTICUS AFTER A 150-DAY GROWOUT IN 0.1-HA PONDS FERTILIZED
WEEKLY WITH CHICKEN LITTER AT 750 KG DRY MATTER/HA
Stocking Final Gross total Net total
Treatment rate weight Survival yield yield Reproduction
Fish/ha g/fish Pct. kg/ha per 150 d
Contro 2,500 352 a 85.0 a 968 c 898 c 221 a
MT-treated 2,500 383 a 88.9 a 1,091 c 1,016 c 243 a
Control 10,000 251 b 69.8 a 1,730 b 1,438 b 10 b
MT-treated 10,000 228 b 92.1 a 2,105 b 1,805 b 12 b
Control 20,000 145 c 83.6 a 2,428 a 1,812 a 0 b
MT-treated 20,000 151 c 85.6 a 2,570 a 1,963 a 4 b
1 Manually-selected male fish.
abc Means within columns followed by different letters are significantly different
(P<0.05).
fish averaged 51.3 percent males, not significantly different
from the expected 1:1 male:female ratio. MT-treated fish
averaged 96.8 percent males. After 94 days, gross tilapia
yield averaged 1,548 kg/ha and 1,613 kg/ha for the control
and MT-treated treatments, respectively.
During growout, slopes of growth curves for con-
trol males and MT-treated males within each stocking rate
were not significantly different except at the low stocking
rate, where MT-treated fish grew faster (Figure 16). How-
ever, at harvest average fish weights for control and MT-
treated males, within each stocking rate, were not signifi-
cantly different. No significant differences in production
data were detected between treatments within stocking rates
(Table 27). The treatment by stocking rate interaction was
not significant.
Growth of control and MT-treated Nile tilapia was
similar. No anabolic response to MT treatment was detected
during the hormone-treatment phase, or during any produc-
tion phase for 244 days subsequent to MT-treatment. The
use of MT for masculinization of tilapia fry was efficacious
and resulted in populations comprised of 97 percent males.
Thus, farmers using semi-intensive culture techniques should
not expect an anabolic response subsequent to sex reversal.
STUDY E4. TILAPIA YIELD IMPROVEMENT BY
MAINTAINING CRITICAL OXYGEN
CONCENTRATIONS IN PONDS
Tilapia endure low dissolved oxygen concentra-
tion by rising to the pond surface to pass oxygen-rich water
across their gills. At El Carao, tilapia have been documented
to endure dissolved oxygen concentrations near 0 mg/L for
up to six hours. Stress from chronic hypoxia was suspected
Nine 0.1-ha ponds were randomly assigned to three
treatments. Vertical pump aerators (0.5 HP AIR-O-LATOR) in
six ponds were activated when DO reached 30 percent or 10
percent of saturation. There was no aeration in the remaining
three ponds. Individual aerators were activated automati-
cally by a computerized data-logging system. Aerators were
operated for one-hour intervals until pond DO exceeded the
critical level.
Ponds were stocked on 8 August 1990 with 24-g
male tilapia at 20,000/ha, 1-g guapote tigre (Cichlasoma
managuense) at 500/ha, and 12-g Cichlasoma maculicauda
at 290/ha. The maculicauda was a native cichlid being tested
for potential as pond culture fish (Study D3). Ponds were
fertilized only with chicken litter at 1,000 kg DM/ha per
week for the first two months of growth. Thereafter, fish
were fed a commercial shrimp ration (20 percent crude
protein) six days per week at 3 percent of tilapia biomass in
the fastest growing treatment. All ponds received equal
quantities of inputs. Duration of the experiment was 148
days.
Aeration resulted in significantly greater fish gross
yield and larger fish than no aeration, but there were no
differences in fish yield between levels of aeration (Table
28). Beneficial effect of aeration was most evident during
the last six weeks (Figure 17). These data indicate that
aeration was not necessary until the end of the experiment
when high inputs resulted in prolonged periods of low DO.
Aerators in both aerated treatments began functioning dur-
ing the first month of growth. Yields among non-aerated
ponds were much more variable than yields among aerated
ponds, indicating that low-oxygen stress may contribute to
the large variation within treatments in tilapia production.
Minimal use of aerators can be used to increase tilapia
production under conditions of low DO, however, such use
must be economically justified.
28
to reduce tilapia growth, even in the
absence of fish mortality. Supplemen-
tal aeration could maintain DO levels
above those detrimental to tilapia
growth. However, the critical oxygen
concentration at which to begin aera-
tion was not known. Indiscriminate
use of aerators increases operating
costs. The objectives of this study
were to determine the critical DO level
at which to initiate aeration in order to
minimize operational time, and to
determine if aeration affected primary
productivity and other water quality
variables.
TABLE 28. MEAN (? S.E.) YIELD OF OREOCHROMIS NILOTICUS, CICHLASOMA MANAGUENSE
(GUAPOTE) AND CICHLASOMA MACULIACUADA AFTER 148 DAYS IN PONDS WITHOUT OXYGEN
REGULATION (CONTROL), OR WITH MINIMUM DISSOLVED OXYGEN CONCENTRATION
MAINTAINED AT 10 PERCENT OR 30 PERCENT OF SATURATION BY AERATION.
Tilapia Maculicauda Guapote Tilapia Gross fish Mean tilapia Tilapia
Treatment yield yield yield reproduction yield weight survival
kg/ha g/fish Pct.
Control 3,404 + 383 8 + 4 20 ? 6 41 + 16 3,473 ? 404 194 15.5 87 ? 3.1
10% 4,133 + 130 15 ? 0 18 + 1 34 + 11 4,201 ? 139 229 2.7 90 ? 2.1
30% 4,269 + 176 19 + 4 19 ? 4 32 ? 1 4,340 ? 182 235 3.9 91 ? 2.3
TABLE 29. MEANS (? S.E.) OF SELECTED WATER QUALITY VARIABLES IN PONDS WITHOUT
OXYGEN REGULATION (CONTROL), OR WITH MINIMUM DISSOLVED OXYGEN CONCENTRATIONS
MAINTAINED AT 10 PERCENT OR 30 PERCENT OF SATURATION BY AERATION
Secchi disk Total Suspended Volatile
Treatment Chlorophyll a visibility Organic N ammonia solids solids
gg/L cm mg NH
3
-N/L mg/L
Control 274 ? 7.3 16.4 ? 0.11 3.43 ? 0.03 0.074 ? 0.013 433 ? 15 167 ? 10
10% 322 ? 17.9 13.3 ? 0.20 3.49 ? 0.02 0.117 ? 0.015 492 ? 9 180 ? 12
30% 342 ? 53.8 12.5 ? 0.94 3.41 ? 0.15 0.122 ? 0.012 538 ? 33 174 ? 8
I Days
Figure 17. Mean weight of Nile tilapia during 140 days of
culture in 0.1-ha earthen ponds without aeration, or with
aeration starting at 10 percent or 30 percent of dissolved
oxygen saturation.
Aeration resulted in small, but significant increases
in total suspended solids (clay turbidity) and total ammonia-
N, but there were no significant differences for organic-N,
chlorophyll a , or total volatile solids (Table 29).
ON-FARM TESTING OF PD/A CRSP FISH
PRODUCTION SYSTEMS
The goal of PD/A CRSP aquacultural research has
been to increase fish production and profitability for small-
and medium-scale commercial producers by using technol-
ogy that enhances natural productivity of ponds with locally
available nutrient inputs.
PD/A CRSP research
results have been dis-
seminated at local, re-
gional, and international
scientific meetings, in
regular lectures at local
vocational-agricultural
schools, at technology-
transfer days at El Carao,
through formulation of
pond management plans
for producers who buy
fingerlings at El Carao,
and in scientific publi-
cations. The next step
was to transfer the pro-
duction technologies to
the farmer. On-farm test-
ing of production sys-
tems would validate research findings, and serve as a teach-
ing tool for extensionists.
In early 1991, the Honduras PD/A CRSP team
developed and implemented a program that linked produc-
ers in the northern and central regions of Honduras with PD/
A CRSP technologies. Participating small- to medium-scale
commercial producers were interested in maximizing prof-
itability by refining their production technology. This group
of farmers participated in the field trials of PD/A CRSP
production systems developed in Honduras and Thailand.
Staff of another technology transfer program, the
USAID/Honduras and MNR Land Use and Productivity
Enhancement (LUPE) project collaborated with PD/A CRSP
in promoting aquacultural development. The LUPE pro-
gram worked with hillside farmers to promote watershed
conservation and sustainable agriculture in the southern and
central regions of Honduras. Many of the farmers that
participated in the LUPE project had few resources such as
fertilizers available to them; use of compost as a fish pond
nutrient, as developed in the Rwanda PD/A CRSP, could be
tested in these hillside ponds.
Peace Corps/Honduras had an on-going fish cul-
ture project that placed Peace Corps volunteers (PCVs) with
MNR. The goal of Peace Corps/Honduras Fish Culture
Project was to improve the economic and nutritional status
of the resource-limited rural population in Honduras through
sustainable fish culture production. Thus, PCVs would test
PD/A CRSP fish production systems on private farms pro-
vided the implementing farmers were interested and had
access to production inputs.
29
Mean individual weight (g/fish)
250 -
O No aeration
225 -
225 O 
10% saturation
200 -
200 A 
30% saturation
175 -
150
125
0
I nn.--~-.--~-- .-1 ~_
Figure 18. Map of Honduras that shows approximate locations of major cities and of farms that participated in the PD/A CRSP
on-farm trials.
PROJECT DESIGN AND IMPLEMENTATION
The two components of this activity were on-farm
trials and short courses in aquaculture for participant
extensionists and farmers. An initial short course preceded
the farm trials. A second seminar was offered upon comple-
tion of trials to summarize and discuss trial results. Hondu-
ran PD/A CRSP personnel were responsible for identifica-
tion and selection of small- and medium-scale commercial
fish farmers to participate in the on-farm trials. Participant
farmers each assigned two ponds for use in trials, which
allowed two production systems to be compared. Each
participating farmer and a PD/A CRSP representative
signed a contract that stipulated the responsibilities of each
party. MNR extension personnel associated with PD/A
CRSP made monthly visits to participant farmers to collect
data, water samples and to provide technical assistance.
LUPE extension personnel identified farmers par-
ticipating in the LUPE project to participate in on-farm
trials. Only one pond per farm would be used to test PD/A
CRSP technologies. LUPE extension personnel were re-
sponsible for supervising data collection by and providing
regular technical assistance. A Honduran PD/A CRSP
aquaculture specialist accompanied LUPE extension
agents on farm visits at least once each month. During
each visit, the PD/A CRSP aquaculture specialist would
collect data, water samples and provide technical assistance.
Peace Corps/Honduras was only collaterally in-
volved in testing PD/A CRSP production systems on farm.
PCVs were responsible for collecting production data at
each of their sites, but they were not obligated to test PD/A
CRSP fish production systems. In practice, greatly different
situations made it difficult to standardize inputs and man-
agement systems.
In July 1991, a one week short course in aquacul-
ture was presented to 17 participating fish farmers and
extensionists in order to provide them with a more thorough
understanding of aquaculture and the on-farm trials in which
they were about to participate. In addition, a producer from
La Villa, Comayagua, and four DIGEPESCA employees
from Comayagua attended the Thursday session on produc-
tion systems, economics and the field trials. Course partici-
pants were provided with reference materials and a Secchi
disk.
30
THREE PRODUCTION SYSTEMS TESTED ON-FARM WERE:
CHEMICAL FERTILIZATION
Nitrogen and phosphorus, as
chemical fertilizer, were the only nutri-
ent inputs to ponds. Nitrogen and phos-
phorus were applied weekly at 30 kg/ha
and 8 kg/ha, respectively. Ammonium
phosphate (18-46-0) and urea (46-0-0)
were the commonly available chemical
fertilizers, therefore weekly fertiliza-
tion rates were 40 kg/ha and 49.5 kg/ha,
respectively. The weekly fertilizer dose
was divided in half, with a three-day
interval between application of each
half dose. Fertilizer was dissolved in
bucket of pond water, which was then
dispersed over the pond surface.
Ponds were stocked with male
Oreochromis niloticus fingerlings at
20,000/ha, and fingerling Cichlasoma
managuense at 500/ha to control tilapia
reproduction.
ORGANIC FERTILIZATION
Animal manure was the only
pond nutrient input. Fresh cattle or swine
manure was mixed with pond water and
the manure slurry was applied to ponds;
chicken litter, when used, was broad-
cast over the pond surface. Manure ap-
plication rate was 500 kg/ha per week
on a dry matter basis.
Ponds were stocked with male
Oreochromis niloticus fingerlings at
10,000/ha and fingerling Cichlasoma
managuense at 500/ha to control tilapia
reproduction.
An alternative fertilization re-
gime increased weekly manure applica-
tions to 750 kg/ha and included urea (25
kg/ha weekly) application. Urea dose
was divided in half, with a three-day
interval between application of each
half dose. Tilapia stocking rate was
increased to 20,000/ha.
FERTILIZATION FOLLOWED BY FEED
Chicken litter was applied to
ponds at a rate of 750 kg dry matter/ha
per week. Urea was also applied weekly
at 25 kg/ha; the weekly dose was di-
vided in half, with a three-day interval
between application of each half dose.
Both fertilizers were applied during the
first 12 weeks of growout, after which
fertilization was suspended and feeding
was initiated. Fish were fed a commer-
cial, pelleted fish feed (25 percent pro-
tein) daily at a rate of 3 percent of fish
biomass. Daily feeding rate was ad-
justed monthly based on average fish
weight, determined by seine sample,
assuming 100 percent fish survival.
Daily feed allowance did not exceed
100 kg/ha.
Ponds were stocked with male
Oreochromis niloticus fingerlings at
20,000/ha, and fingerling Cichlasoma
managuense at 500/ha to control tilapia
reproduction.
The PD/A CRSP, through the El Carao National
Fish Culture Research Center in Comayagua, provided at no
cost the tilapia and guapote fingerlings necessary to stock
ponds. All other production costs were borne by the farmer.
Field data were collated and analyzed by PD/A CRSP
personnel at El Carao. Production trials were to last 150
days. Prior to stocking ponds, total alkalinity in pond water
was analyzed to determine the need for lime; liming was
considered unnecessary if total alkalinity exceeded 20 mg/
L as CaCO
3
.
A total of 13 small- and medium-scale commercial
fish farmers was selected to participate in the on-farm trials
through direct linkage with El Carao (Figure 18). Most
producers tested two production systems on their farm:
chemicalfertilization and fertilization followed byfeed. Ten
farmers completed the trials but only seven adhered to the
management system. Two farmers initiated a second series
of trials, repeating the fertilization followed by feed at
stocking rates of 20,000 to 30,000/ha.
LUPE extension personnel selected seven farmers
to participate in the trials in September 1991 (Figure 18).
Farmers collected animal manures for pond nutrient input.
Only one farmer completed the trial; supervision and data
collection at all farms suffered because of severe administra-
tive problems in the LUPE project.
Peace Corps Volunteers worked with 13 client
farmers (Figure 18). Farmers did not test PD/A CRSP pro-
duction systems, but rather used a variety of fish culture
practices, with nutrient inputs such as termite nests, manure,
wheat bran, rice-polishing, and waste or cracked corn. Ponds
were often co-stocked with other fish species, especially
tambaquf (Colossoma macropomum); Cichlasoma
managuense, to control tilapia reproduction, was stocked by
less than 20 percent of the farmers.
RESULTS OF ON-FARM TRIALS
SMALL TO MEDIUM SCALE FARMERS
Mean fish yields after 157 days were 2,413 and
1,785 kg/ha for the fertilization followed by feed and chemi-
cal fertilization systems, respectively (Table 30). These
means included data from all producers even though devia-
tion from the work plan was suspected when water quality
measurements, e. g., ammonia and Secchi disk visibility,
31
I
were below ex-
pected values for
the management
system being
tested. The pre-
dominant reason
for noncompli-
ance appeared to
be that the farm
owner was not
directly involved
in daily pond
management, but
rather relied on a
farm manager
who did not
share the owner' s
commitment to
the trials.
M e a n
yields increased
to 2,890 and
2,180 kg/ha in
162 days for the
fertilization fol-
lowed by feed and
fertilization systems
tively, when data fi
complying farmers
cluded from the
(Table 30). Mean fir
size also increased
(Table 30); tilapia o
were marketable in
ras. The minimum
Honduran consun
cepted was 100 to 1
smaller than this w
cult to market in urt
but reports from ru
chemical
s, respec-
rom non-
were ex-
analysis
ial tilapia
I slightly
f this size
SHondu-
size fish
ners ac-
25 g. Fish
ere diffi-
an areas,
iral areas
indicated that it was possible to sell fish as small as 50 g.
In each treatment, tilapia mean weight at each
sampling was regressed against time to obtain a general
indication of growth across farms (Figure 19). The regres-
sion equations were:
Y = 0.803 X + 5.859, R
2 
= 0.661, forfertilization
followed by feed, and
Y = 0.639 X + 9.949, R
2 
= 0.626, for chemical
fertilization,
where Y equals mean individual weight (g/fish) and X
equals day number. Observed fish growth was about 25
percent higher in the fertilization followed by feed system.
The variability observed in the data can be attributed to
noncompliance to the work plan, stocking rate differences,
and site differences.
Results of water quality analyses are shown in
Table 31. Mean total ammonia concentrations and pH were
similar in both treatments. Observed mean total alkalinity
was greater in the fertilization followed by feed system.
The range of yields obtained by farmers (Table 30)
were comparable to the range of yields obtained on the
Experiment Station. The total quantity of inputs used by
32
TABLE 30. ON-FARM TRIAL PRODUCTION RESULTS FOR SMALL TO MEDIUM SCALE TILAPIA FARMERS
Date Stocking Mean fish Gross
Producer Treatment
1  
Pond area stocked rate Duration weight yield Survival
m
2  
fish/ha days g/fish kg/ha Pct.
Barleta Fert. + feed 2,400 16 Apr. 91 30,000 163 154 2,373 51
Castellon Fert. + feed 2,400 23 Aug. 91 20,000 152 153 2,429 79
EAP Fert. + feed 900 4 Oct. 91 18,000 180 215 3,173 82
Ferrera Fert. + feed 230 21 May 91 30,000 176 108 3,149 98
Funez
2  
Fert. + feed 234 15 May 91 20,000 146 75 1,457 97
Garcia Fert. + feed 650 4 June 91 20,000 148 204 3,838 94
INA
2  
Fert. + feed 875 15 May 91 20,000 146 108 2,163 85
Lobo
2  
Fert. + feed 200 18 June 91 20,000 153 64 756 59
Rodriguez Fert. + feed 800 7 June 91 20,000 161 192 2,381 62
Means all farmers 965 22,000 158 141 2,413 79
compliant farmers 1,230 23,000 163 171 2,890 78
Barleta Chemical 2,400 16 Apr. 91 30,000 163 58 1,591 91
Castellon Chemical 2,400 23 Aug. 91 20,000 152 117 2,153 92
EAP Chemical 900 4 Oct. 91 18,000 180 182 3,054 93
Funez
2  
Chemical 234 15 May 91 20,000 146 97 1,531 79
Garcia Chemical 1,250 4 June 91 20,000 148 154 1,511 49
INA
2  
Chemical 875 15 May 91 20,000 146 59 1,060 86
Lobo
2  
Chemical 200 18 June 91 20,000 153 63 790 63
Rodriguez Chemical 4,837 7 May 91 17,000 161 170 2,588 80
Means all farmers 1,650 20,625 156 113 1,785 79
compliant farmers 2,377 21,000 161 136 2,180 81
I Fert. + feed: Organic fertilizer then feed; Chemical: chemical fertilization only; see text for details.
2 Farmers known to have deviated 
from work plan.
TABLE 31. MEAN WATER QUALITY VARIABLES IN PONDS ON SMALL- TO MEDIUM-
SCALE TILAPIA FARMS DURING ON-FARM TRIALS
Total alkalinity Total ammonia
Farmer Feed and fert. Chemical Feed and Fert. Chemical Feed and fert. Chemical
mg/L as CaCO
3  
pH mg/L NH
3
-N
Barleta 68 145 10.0 9.0 0.20 0.38
Castellon 56 82 9.5 10.0 0.42 0.12
Ferrera - 56 - 8.3 - 0.07
Funez 86 120 8.5 8.0 0.02 0.16
Garcia 103 120 8.7 8.0 0.67 0.44
INA 95 112 10.0 8.0 0.16 0.07
Lobo 39 47 7.0 6.8 0.07 0.05
Rodriguez 86 137 9.0 8.5 0.47 0.43
Mean 76 102 7.8 7.6 0.29 0.21
Mean individual weight
(g/fish)
225 - Barleta FERTILIZER THEN 
FEED
Castellon
200 - INA 13
0 Ferrera
175 - i Funez
o Garcia A
150 - A Lobo
S A Rodriguez
125 - A
0' A ~ i
100
75
50
25
0
A
w
A
Mean individual weight
(g/fish)
180 - > Barleta 
CHEMICAL FERTILIZATION
S Castellon A
160- INA
1 Funez
O
O
" O
I l I
I I I I I I I I
0 20 40
I I
60 80 100 120 140 160 180
Days
140
120
100
80
60
40
20
0
- Q Garcia 
A
A Lobo
- A Rodriguez
- c A
==3
0 20 40 60 80 100 120 140 160 180
Days
Figure 19. Growth of tilapia during on-farm conducted in collaboration with small to medium scale commercial tilapia farmers
in Honduras. Management systems used were: fertilization followed by feed (left) and chemical fertilization (right).
each farmer during the trials is shown in Table 32. Our
observation was that farmers who adhered strictly to the
work plan obtained the greatest yields, barring unexpected
TABLE 32. INPUTs USED DURING ON-FARM TRIALS BY
SMALL- TO MEDIUM-SCALE TILAPIA FARMERS
Chemical fertilizer 
Chicken
Producer Treatment Urea 18-46-0 litter
kg kg kg
Barleta Fertilization then feed 60 0 2,422
Castellon Fertilization then feed 60 0 1,995
EAP Fertilization then feed 23 0 1,640
Ferrera Fertilization then feed 9 0 234
Funez' Fertilization then feed
Garcia Fertilization then feed 16 0 1,097
INA' Fertilization then feed
Lobo' Fertilization then feed 592
Rodriguez Fertilization then feed 24 0 738
Means all farmers 28 0 1,180
compliant farmers 34 0 1,297
Barleta Chemical fertilization 337 250 0
Castellon Chemical fertilization 290 218 0
EAP Chemical fertilization 156 206* 0
Funez' Chemical fertilization
Garcia Chemical fertilization 145 109 0
INA' Chemical fertilization
Lobo' Chemical fertilization 27 0 0
Rodriguez Chemical fertilization 610 448 0
Means all farmers 282 205 0
compliant farmers 345 256 0
Producers known to have deviated sharply from standardized wor
* Fertilizer 0-20-0.
fish mortality.
Partial enterprise budgets were developed for each
production system based on mean data from farmers who
complied with the work plan (Table 33). Average
income above input costs was $930 and $407/ha per
162-day growout cycle (US$1= Lempiras 5.40, July
1992) for thefertilizationfollowed byfeed and chemi-
Feed calfertilization 
systems, respectively. A 
farmer that
kg utilized the fertilization 
followed by feed systemkg
196 would need 
to have a greater 
amount of available
196
635 capital, 
and would have more 
capital at risk.
574
72 Results of the second trials showed that in-
218 creasing the fish 
stocking rate from two 
to three/m
2
did not increase total yields at either of the two farms.
0 Gross tilapia yield at Ferrera's farm was 2,866 kg/ha
340 in 122 days. After 167 days tilapia yields at Garcia's
243 farm 
were 3,779 
and 3,259 kg/ha 
for the two 
fish/m
2
292 and three fish/m
2 
stocking rates, respectively. Feed
conversion efficiency at one of the farms indicated
0 that increased density was 
not accompanied by a0
0 proportionate increase in 
feed use. Efficiency was
1.03 kg feed/kg gain for the higher density and 1.50
0 at the lower density at Garcia's farm. A 
reluctance to
0 increase feed 
inputs necessary 
for good growth 
re-
0 sulted in small, marginally marketable fish at the
0 higher density. Income above variable 
costs was less
0 at the higher density.
33
II
) GA
LUPE FARMERS
The LUPE farmer who completed the farm
trial harvested 1,048 kg/ha of tilapia after 207 days. This
farmer made weekly applications of fresh cow manure at
TABLE 33. PARTIAL ENTERPRISE BUDGET PER 1 HA POND FOR Two DIFFERENT POND
MANAGEMENT STRATEGIES USED BY SMALL- TO MEDIUM-SCALE COMMERCIAL FARMERS
DURING ON-FARM TRIALS. VALUES IN HONDURAN LEMPIRAS (L.5.40 = $1 U.S.)
Fertilization Chemical
plus feed fertilization
Description Unit cost Unit Quantity Cash Quantity Cash
Income
adult tilapia 7.17 kg 2,890 20,721 2,180 15,631
Total Income 20,721 15,631
Variable costs
fingerlings
tilapia 0.10 each 23,000 2,300 21,000 2,100
guapote 0.15 each 500 75 500 75
plastic bags 6.00 each 47 282 43 258
feed (20% protein) 76.70 45-kg sack 53 4,065 0 0
fertilizer
chicken litter 2.28 27-kg sack 400 912 0 0
urea 69.00 45-kg sack 7 483 33 2,277
18-46-0 82.00 45-kg sack 0 0 24 1,968
transport
fingerlings 200.00 60-km trip 1 200 1 200
feed 200.00 60-km trip 1 200 0 0
chicken litter 200.00 60-km trip 2 400 0 0
fertilizer 200.00 60-km trip 0 0 1 200
field labor
day 14.00 day 87 1,218 72.5 1,015
night 24.50 night 162 3,969 162 3,962
irrigation water 25.00 ha-m 2 50 2 50
interest on variable
capital 
0.23 year 
0.5 1,544 
0.5 1,321
Total Variable Costs 15,698 13,433
Income Above
Variable costs 5,024 2,198
had ceased be-
fore fish were
harvested. Thus,
factors of pro-
duction (ponds,
inputs, farmer's
labor, etc.) were
underutilized.
Little eco-
nomic data were
available for this
group of farmers.
Few inputs were
purchased, and it
was not possible
to assign an ac-
curate economic
value to inputs
collected from
the field, e. g.,
manure, termite nests, etc. Thus, it was impossible to con-
duct an economic evaluation of the traditional fish produc-
tion system.
34
only 100 kg/ha (dry matter basis);
and weekly urea fertilization of 20
kg/ha was suspended during the fi-
nal six weeks.
PEACE CORPS FARMERS
Mean tilapia yield was 1,343 kg/
ha after an average grow-out period
of 211 days; other fish yield aver-
aged 117 kg/ha, for a gross fish yield
of 1,484 kg/ha (Table 34). Fish yield
varied from 315 to 3,163 kg/ha. In
discussions with the PCVs it be-
came apparent that fish yield varied
in relation to the amount of inputs
used, i. e., adherence to a manage-
ment plan. Farmers that consistently
provided nutrient inputs to the pond
obtained yields that approximated
those achieved by the small- and
medium-scale commercial farmers.
Tilapia size at harvest averaged 122
g, similar to sizes obtained by small/
medium scale farmers. Lack of data
did not allow tilapia growth to be
characterized, but experience sug-
gests that the grow-out period was
too long, that significant fish growth
TABLE 34. PRODUCTION RESULTS FROM PEACE CORPS VOLUNTEER ASSISTED
TRADITIONALLY-MANAGED FISH PONDS
Pond Date Stocking Mean Tilapia Other Gross Tilapia
Producer area stocked rate Duration tilapia yield fish yield survival
weight yield
m
2  
fish/ha Days g/fish kg/ha kg/ha kg/ha Pct.
Canahuati 600 27 Aug. 91 22,500 
* * * * * 
*
Canahuati 600 27 Aug. 91 22,500 
* * * * * 
*
Escobar 235 20 Nov. 91 4,468 145 90 314 0 314 
78
Exito Campesino 1,500 23 Oct. 91 23,300 * * * * 
* 86
Hernandez 150 * * * 55 1,093 0 1,093 
*
La Escuela 330 * * 145 104 523 41 564 
*
Leba 395 11 May 91 8,481 339 168 1,382 0 1,382 
97
Miranda 325 5 Nov. 91 12,308 162 80 862 0 862 87
Mercado 350 * * * 109 785 441 1,226 
34
Rosa Urbana 480 22 Aug. 91 20,167 234 183 2,249 229 
2,479 61
Villa Nueva 80 * * * 93 1,116 0 1,116 
*
Zelaya 1,000 22 Aug. 91 25,000 235 146 2,321 319 
2,640 64
Zelaya 350 Mar. 91 20,000 237 195 2,785 377 3,163 
71
Means 492 17,636 214 122 1,343 117 1,484 
72
COMPARISON OF PRODUCTION SYSTEMS
In Honduras, pond management practices charac-
terized by varied and sporadic nutrient inputs and growout
cycles that are considerably longer than 150 days can be
collectively considered as the "traditional production sys-
tem." This includes non-compliant small- and medium-
scale commercial farmers and subsistence farmers. These
producers provide a basis against which to compare the
impact of the PD/A CRSP production systems. Twofold
increases in tilapia yield relative to the traditional system
were obtained by farmers using PD/A CRSP production
systems (Table 35). Other important differences between the
TABLE 35. COMPARISON OF SELECTED PRODUCTION PARAMETERS BETWEEN THE TRADITIONAL
SYSTEM AND THE FERTILIZATION FOLLOWED BY FEED AND CHEMICAL FERTILIZATION SYSTEMS
USED BY FISH CULTURISTS DURING ON-FARM TRIALS IN HONDURAS
Pct. difference Pct. difference Pct. difference
Stocking relative to Grow-out relative to Annual relative to
Production system rate traditional duration traditional tilapia yield traditional
system system system
Fish/m2 Days kg/ha/yr.
Traditional system 1.9 -- 189 -- 
2,579 ---
Fert. followed by feed
compliant farmers 2.3 +21 163 -14 6,496 +152
Chemical fertilization
compliant farmers 2.1 +11 161 -15 4,904 +90
demonstrated that the PD/
A CRSP technologies were more productive than the tradi-
tional tilapia production system used in Honduras. The
limited enterprise budget analysis indicated that both PD/A
CRSP systems resulted in significant income above variable
costs, an indicator of the economic viability of the systems.
It should be noted that the PD/A CRSP production sys-
tems tested in this trial were not developed for subsis-
tence fish farmers, but rather for small- to medium-scale
commercial fish farmers who have the capability to
purchase the necessary inputs. It is this group of fish
farmers whom we feel will have the greatest impact on
freshwater aquaculture in Honduras.
EcoNoMlIcs OF IHIoNDURAS IPID/A CRSIP
POND MANAGEMENT SYSTEMS
Profitability of 41 production systems having dif-
fering nutrient input regimes in Honduras between 1983 to
1992 was evaluated by enterprise budget analysis. Enter-
prise budget items for the various treatments are discussed
and net returns were calculated using prices and financial
conditions in Honduras for 1992.
Positive net returns in U.S. Dollars (Lempiras 5.4
= U.S.$) and return on investment (percentages) demon-
strated the variety of viable enterprise choices available to
Honduran fish farmers under varying economic conditions.
Fish yields were dependent on input type and quantity.
Various inputs were transformed through production yields
and associated infrastructural needs into cash receipts,
input costs and fixed costs. Income above variable cost
calculation was significant because it represented short-
term viability of the operation. Positive income above vari-
able costs is a short term measure indicating that operations
should continue even with an overall loss (net returns)
because shutting down operations would only increase over-
all loss. Long term viability of the operation though would
need to have positive net returns. As interest rates, prices
and availability of inputs change, a knowledgeable farmer
recognizes that changing production systems could posi-
tively affect profits. An analysis of profitability for the 41
different tilapia production systems could assist the farmer
in making economic decisions.
TREATMENT CATEGORIES
Five categories of nutrient input regimes were
researched in Honduras: (1) Chemical Fertilization ; (2)
Organic Fertilization; (3) Organic plus Chemical Fertiliza-
tion; (4) Organic Fertilization plus Supplemental Feed; and
(5) Feed Only (Table 36). Stocking rates varied from 2,500
to 30,000 male Oreochromis niloticus /ha of pond surface
area. Production inputs centered around the choice of chemi-
cal fertilizers, organic fertilizers, formulated diets, combina-
tions thereof, stocking rates, use of aeration and season.
Prices for these production inputs may vary over time and
local conditions. Chemical fertilizer prices can vary depend-
ing on their local availability and the degree that other
35
traditionaland PD/ACRSP
systems were that with the
traditional system stocking
rate was about 16 percent
lower and the grow-out du-
ration was about 15 per-
cent longer (Table 35).
SUMMARY
Results of on-farm tri-
als with the small- and me-
dium-scale tilapia farmers
TABLE 36. NUTRIENT INPUT CATEGORIES, TREATMENT NAMES, STOCKING RATES, NUTRIENT INPUTS AND
STUDY NUMBER FOR HONDURAS PD/A CRSP RESEARCH, 1983 TO 1992
Category Treatment Stocking rate' Nutrient inputs
Chemical fertilization
Organic fertilization
Chem fert/cycle I
Chem fert/cycle II
N+P/C
N+P/W
Cow manure
2,500/ha + CL
10,000/ha + CL
20,000/ha + CL
CL 125
CL 250
CL 500
CL 1,000
CL 1,000 @ 2
CL 750/no urea
CL 750/no sub
C:N control
CL 500/no feed
10,000/ha
10,000/ha
20,000/ha
20,000/ha
10,000/ha
2,500/ha
10,000/ha
20,000/ha
10,000/ha
10,000/ha
10,000/ha
10,000/ha
20,000/ha
10,000/ha
10,000/ha
20,000/ha
10,000/ha
8.7 kg TSP
2
/ha every 2 weeks; Study B1
Weekly: 30.6 kg Urea/ha + 62.6 kg TSP/ha; Study B2
Weekly: 40.6 kg Urea/ha + 31.0 kg DAP
3
/ha; cool season control; Study B7
Weekly: 39.5 kg Urea/ha + 26.8 kg DAP/ha; warm season control; Study B7
Weekly: 1,020 kg Cow manure/ha, dry matter basis; Study B7
Weekly: 750 kg CL
4
/ha; Study E3
Weekly: 750 kg CL/ha; Study E3
Weekly: 750 kg CL/ha; Study E3
Weekly: 125 kg CL/ha; Study B3
Weekly: 250 kg CL/ha; Study B3
Weekly: 500 kg CL/ha; Study B3
Weekly: 1,000 kg CL/ha; Study B3
Weekly: 1,000 kg CL/ha; Study B4
Weekly: 750 kg CL/ha; control for CL 750/urea; Study B5
Weekly: 750 kg CL/ha; control for CL 750/x sub; Study C4
Weekly: 750 kg CL/ha; control for Cx:N; Study B6
Weekly: 500 kg CL/ha; control for CL ~00/x% feed; Study C3
Organic plus chemical fertilization
CL 750/Urea
C8:N1
C6:N1
C4:N1
CL 750/N+P/C
CL 500/N+P/C
CL 250/N+P/C
CL 500/N+P/W
CL 250/N+P/W
10,000/ha Weekly: 750 kg CL/ha + 10 kg N/ha as urea; Study B5
20,000/ha Weekly: 750 kg CL/ha + 6.3 kg N/ha as urea; Study B6
20,000/ha Weekly: 750 kg CL/ha + 14.1 kg N/ha as urea; Study B6
20,000/ha Weekly: 750 kg CL/ha + 29.9 kg N/ha as urea; Study B6
20,000/ha Weekly: 750 kg CL/ha + 29.7 kg Urea/ha; cool season; Study B7
20,000/ha Weekly: 500 kg CL/ha + 37.5 kg Urea/ha; cool season; Study B7
20,000/ha Weekly: 250 kg CL/ha + 41.3 kg Urea/ha + 9.9 kg DAP/ha; cool season;
Study B7
20,000/ha Weekly: 500 kg CL/ha 33.5 kg Urea/ha + 9.5 kg DAP/ha; warm 
season;
Study B7
20,000/ha Weekly: 250 kg CL/ha 37.6 kg Urea/ha + 15.4 kg DAP/ha; warm
season; Study B7
Organic fertilization plus supplemental feed
CL 500/0.5% feed 10,000/ha Weekly: 500 kg CL/ha. Daily: feed (0.5% fish biomass); 
Study C3
CL 500/1.0% feed 10,000/ha Weekly: 500 kg CL/ha. Daily: feed (1.0% fish biomass); 
Study C3
CL 500/2.0% feed 10,000/ha Weekly: 500 kg CL/ha. Daily: feed (2.0% fish biomass); 
Study C3
CL 750/1 mon. sub. 10,000/ha Weekly (first 30 d): 750 kg CL/ha. 
Daily (begin. day 31): feed (3.0%
fish biomass); Study C4
CL 750/2 mon. sub. 10,000/ha Weekly (first 60 d): 750 kg CL/ha. Daily 
(begin. day 61): feed (3.0%
fish biomass); Study C4
CL 750/3 mon. sub. 10,000/ha Weekly (first 90 d): 750 kg CL/ha. Daily (begin. day 
91): feed (3.0%
fish biomass); Study C4
CL 60/feed 20,000/ha Weekly (first 60 d): 1,000 kg CL/ha. Daily 
(begin. day 61): feed (3.0%
fish biomass); Study C2
CL plus feed 20,000/ha Weekly: 500 kg CL/ha. Daily: feed (1.5% fish biomass); 
Study C2
No aeration/sub. 20,000/ha Weekly (first 60 d): 1,000 kg CL/ha. Daily (begin. 
day 61): feed (3.0%
fish biomass); Study E4
10% sat./sub. 20,000/ha Weekly (first 60 d): 1,000 kg CL/ha. Daily (begin. 
day 61): feed (3.0%
fish biomass); Study E4
30% sat./sub. 20,000/ha Weekly (first 60 d): 1,000 kg CL/ha. Daily (begin. 
day 61): feed (3.0%
fish biomass); Study E4
10,000/ha + feed/C
20,000/ha + feed/W
20,000/ha + feed/C
30,000/ha + feed/C
10,000/ha
20,000/ha
20,000/ha
30,000/ha
Daily: feed (3.0% fish biomass); cool season; Study C1
Daily: feed (3.0% fish biomass); warm season; Study C2
Daily: feed (3.0% fish biomass); cool season; Study Cl
Daily: feed (3.0% fish biomass); cool season; Study C1
'Male Oreochromis niloticus.
2 Triple superphosphate fertilizer (0-46-0).
3 
Diammonium phosphate fertilizer (18-46-0).
4 Chicken litter applied on dry matter basis.
36
Feed only
agricultural users compete for fertilizers. Animal manures
are usually less expensive than chemical fertilizers or
formulated diets. Manure value is usually based upon the
labor of collection plus transportation and profit margin. In
these production trials, chicken litter was bagged and deliv-
ered to the experiment station by a local entrepreneur.
Formulated fish feeds were relatively expensive inputs com-
pared to chemical and organic fertilizers.
ENTERPRISE BUDGET ANALYSIS
BUDGET PREPARATION
Full-cost enterprise budgets included use of 1992
Honduran prices for market value of fish, variable costs and
fixed costs. Use of all variable and fixed inputs, including
credit, was assumed. Table 37 depicts representative enter-
prise budgets for the five nutrient input categories. Cash
receipts were obtained from the sale of adult tilapia, adult C.
managuense and tilapia fingerlings. Variable input items
included: fingerlings, feed, fertilizer, transport of materials,
aeration electricity and interest on operating costs .Quanti-
ties of some variable inputs were similar for all-treatments;
these inputs were: irrigation water, day and night field labor,
and transport of fingerlings (one trip for all treatments
stocking less than 30,000 fish/ha, otherwise two trips were
required). Aeration costs were calculated by multiplying
aerator kilowatt usage by the number of hours of aerator
operation multiplied by the kw/h electrical cost. Fish pro-
duction cycles were approximately five months for all treat-
ments. Enterprise budgets were annualized on a per hectare
basis for comparative purposes.
TABLE 37. REPRESENTATIVE FULL-COST ENTERPRISE BUDGET FOR EACH OF THE FIVE NUTRIENT INPUT
CATEGORIES EVALUATED FOR TILAPIA PRODUCTION PONDS IN HONDURAS. VALUES IN U.S. DOLLARS
PER HECTARE PER SIX-MONTH PRODUCTION CYCLE
Unit Chemical Manure only Manure plus Manure plus Feed only
cost/ fertilization (20,000/ha/ chemical supplemental (feed only)
price only (N+P/W) manure) fertilization feed
(C6:N1) (CL plus feed)
Description Quantity Cash Quantity Cash Quantity Cash Quantity Cash Quantity Cash
Cash receipts
adult tilapia 1.43 2,045 2,924 2,497 3,570 3,640 5,204 4,021 5,749 4,470 6,390
adult guapote 1.43 5 7 53 76 12 17
tilapia repro 0.10 29 3 2 0 12 1 302 31 780 79
Total cash receipts 2,934 3,646 5,222 5,779 6,470
Variable costs
fingerlings
tilapia 0.02 20,000 370 20,000 370 20,000 370 20,000 370 20,000 370
red tilapia 0.03
guapote 0.03 500 14 1,000 28 500 14
plastic bags 1.11 41 46 42 47 41 46 40 44 40 44
feed 12.04 84 1,011 200 2,407
fertilizer
chicken litter 0.37 692 256 694 257 352 130
cow manure 0.00
urea 12.59 19 239 7 88
18-46-0 15.19 13 197
0-46-0 17.04
transport
fingerlings 37.04 1 37 1 37 1 37 1 37 1 37
feed 37.04 1 37 1 37
chicken litter 37.04 4 148 4 148 2 74
cow manure
fertilizer 37.04 1 37
aeration
field labor
day 2.59 75 194 75 194 75 194 75 194 75 194
night 4.54 150 681 150 681 150 681 150 681 150 681
irrigation water 4.63 1 5 1 5 1 5 1 5 1 5
interest on
variable capital 0.23 209 203 212 297 434
Total variable costs 2,030 1,969 2,051 2,881 4,210
Income above variable costs 904 1,677 3,171 2,898 2,260
37
TABLE 37, CONTINUED. REPRESENTATIVE FULL-COST ENTERPRISE BUDGET FOR EACH OF THE FIVE NUTRIENT INPUT
CATEGORIES EVALUATED FOR TILAPIA PRODUCTION PONDS IN HONDURAS. VALUES IN U.S. DOLLARS
PER HECTARE PER SIX-MONTH PRODUCTION CYCLE
Unit Chemical Manure only Manure plus Manure plus Feed only
cost/ fertilization (20,000/ha/ chemical supplemental (feed only)
price only (N+P/W) manure) fertilization feed
(C6:N1) (CL plus feed)
Description Quantity Cash Quantity Cash Quantity Cash Quantity Cash Quantity Cash
Fixed costs
Interest on bldg.
and equipment 0.23 4,939 1,136 4,939 1,136 4,949 1,136 5,310 1,221 4,939 1,136
Depreciation on
bldg. and equip. 139 139 139 164 139
Other 0.28 8 2 8 2 8 2 8 2 8 2
Total fixed costs 1,277 1,277 1,277 1,387 1,277
Total costs 3,307 3,246 3,329 4,268 5,488
Net returns to land
and management
Dollars -373 399 1,893 1,511 982
Dollars/yr. -746 799 3,787 3,022 1,965
Return to average investment 16 77 57 40
In Honduras, the 1992 interest rate for agricultural cost in the determination of economic profitability 
or effi-
loans was 23 percent. Because all trials were conducted at cient resource utilization.
the experiment station no actual borrowing of money oc-
curred, however, farmers may or may not need access to
Fixed costs included depreciation and interest on
credit. The 23 percent interest rate represents an opportunity buildings and equipment, and an annual cost for pond
TABLE 38. BUILDING AND EQUIPMENT DEPRECIATION AND INTEREST CHARGE ON AVERAGE INVESTMENT FOR FIVE NUTRIENT INPUT
CATEGORIES IN HONDURAS 1983 TO 1992. ALL VALUES ARE IN U.S. DOLLARS (U.S. $1 = LEMPIRAS 5.40 IN JULY 1992)
Depreciation Charges (U.S. $)1
Proportion Salvage Useful Chemical Organic Chemical Org. Fert. plus Feed
Item Value charged value life fertilization fertilization plus organic supplemental only
fertilization 
feed
2
U.S. $ Pct. Pct. Yrs.
Storage building
3  
740 100 0 15 49.38 49.38 98.76 98.76 49.38
Pick-up truck
4  
7,407 10 10 10 66.67 66.67 66.67 66.67 66.67
D.O. meter
5  
900 100 10 5 162.00 162.00 162.00 162.00 162.00
Electric aerator 400 100 10 5 --- --- --- [72.00] ---
Pond/harvest basin 11,111 0 100 20 0.00 0.00 0.00 0.00 0.00
Annual depreciation --- --- --- --- 278.05 278.05 327.43 327.43 278.05
[399.43]
Depreciation per 6- --- --- --- --- 139.03 139.03 163.72 163.72 139.03
month production cycle [199.72]
Interest on building and --- --- --- --- 1,136.07 1,136.07 1,221.26 1,221.26 1,221.26
equipment
6  
[1,271.86]
1 Straight-line depreciation was used and calculated as: (Price-Salvage value)/(Useful life).
2 There were two aeration treatments in this category that required one aerator; values in brackets include additional depreciation
and interest charges for the aerator.
3 One storage building for treatments using feed or fertilizer; two storage buildings were required for treatments that combined feed
or fertilizer with manure.
4 Pick-up truck was charged to the fish farm enterprise 10% of 
the time.
5 Dissolved oxygen meter (at U.S. cost plus import dury) was 
included.
6 Interest on building and equipment was calculated as: ((cost + salvage value)/2) 
x 23%.
38
TABLE 39. SUMMARY ECONOMIC RETURNS PER HECTARE PER SIX-MONTH PRODUCTION
CYCLE AND PERFORMANCE FROM ENTERPRISE BUDGET ANALYSES OF TILAPIA PRODUCTION
POND NUTRIENT INPUT REGIMES IN HONDURAS BETWEEN 1983 AND 
1992.
TILAPIA MARKET PRICE IN 1992 WAS U.S. $1.43 PER KILOGRAM.
CURRENCY EXCHANGE RATE AT THE TIME OF ANALYSIS WAS U.S. $1= 5.40 
LEMPIRAS
Breakeven cost per
Income above Net 
returns Return to 
kilogram of fish to cover:
Treatment variable to land and average Variable Variable 
and
costs management investment costs fixed costs
Chemical fertilization
Chem Fert/Cycle I
Chem Fert/Cycle II
N+P/C
N+P/W
Organic fertilization
U.S. $
-679.86
-102.06
562.19*
904.19*
Cow manure 1,071.24*
2,500/ha + CL -306.85
10,000/ha + CL 1,057.87
20,000/ha + CL 1,676.75
CL 125 324.81*
CL 250 818.43
CL 500 1,167.07
CL 1,000 1,507.84
CL 1,000 @ 2 1,082.59*
CL 750/no urea 704.21*
CL 750/no sub 876.29*
C:N control 2,040.56
CL 500/no feed 169.62*
Organic plus chemical fertilization
CL 750/urea 798.88*
C8:N1 1,918.17
C6:N1 3,170.84
C4:N1 1,676.88
CL750/N+P/C 611.70
CL 500/N+P/C 1,465.28
CL 250/N+P/C 1,362.44
CL 500/N+P/W 3,026.02
CL 250/N+P/W 1,302.65
U.S. $
-1,957.00
-1,379.00
-715.00
-373.00
-206.00
-1,584.00
-220.00
399.00
-953.00
-459.00
-110.00
230.00
-195.00
-573.00
-401.00
763.00
-1,108.00
-478.00
641.00
1,893.00
400.00
-666.00
188.00
85.00
1,749.00
25.00
Organic fertilization plus supplemental feed
CL 500/0.5% feed 337.58* -1,050.00
CL 500/1.0% feed 493.81 -893.00
CL 500/2.0% feed 289.75* -1,097.00
CL 750/1 mon. sub. 402.71* -984.00
CL 750/2 mon. sub. 307.02* -1,080.00
CL 750/3 mon. sub. 693.80 -693.00
CL 60/feed 2,798.85 1,412.00
CL plus feed 2,898.06 1,511.00
No aeration/sub. 1,145.96* -241.00
10% sat./sub. 2,134.19 660.00
30% sat./sub. 2,286.11 812.00
Feed only
10,000/ha + feed/C
20,000/ha + feed/W
20,000/ha + feed/C
30,000/ha + feed/C
835.48*
2,259.72
1,023.80*
2,014.59
-442.00
982.00
-254.00
737.00
Pct.
-79
-56
-29
-15
-8
-64
-9
16
-39
-19
-4
9
-8
-23
-16
31
-45
-19
26
77
16
-27
8
3
71
1
-40
-34
-41
-37
-41
-26
53
57
-9
24
29
-18
40
-10
30
U.S. $ U.S. $
2.97
1.51
1.13
0.99
0.77
1.87
0.90
0.79
1.16
0.90
0.82
0.79
0.94
1.01
0.97
0.70
1.29
1.00
0.73
0.56
0.81
1.12
0.83
0.85
0.58
0.87
1.19
1.13
1.27
1.29
1.31
1.14
0.82
0.72
1.10
0.92
0.90
1.08
0.94
1.15
1.01
5.87
2.50
1.82
1.62
1.56
3.48
1.57
1.30
2.25
1.74
1.49
1.34
1.52
1.79
1.69
1.16
2.37
1.71
1.20
0.91
1.28
1.78
1.36
1.40
0.94
1.42
2.18
1.97
2.04
1.88
1.95
1.77
1.12
1.06
1.51
1.28
1.25
1.61
1.23
1.50
1.28
* Income above variable cost was positive when net returns to land and management was
negative. Positive income above variable costs is a short-term measure that indicates that
operations should be continued even with an overall loss because shutting down operations
would increase the overall loss. Long-term viability of the operation would require positive
net returns.
maintenance (Table 38). Calcula-
tions included one storage build-
ing for treatments using feed or
fertilizer and two storage build-
ings when feed or chemical fertil-
izer was used in combination with
manure to store inputs separately.
One truck was charged to the fish
farming enterprise 10 percent of
the time and one dissolved oxy-
gen meter was included (at U.S.
cost plus import duty). Buildings
and equipment were charged at
the prevailing 23 percent interest
rate using a straight line method
of depreciation and the same rate
was applied on average invest-
ment items. It was assumed the
land was already owned and there
were existing ponds so no costs
were included for these items.
Ponds were not depreciated. Sub-
tracting variable and fixed costs
from the cash receipts resulted in
a net return to land, existing ponds
and operator's labor and manage-
ment. This is referred to as "net
returns" in discussion hereafter
unless specifically stated other-
wise.
Breakeven cost is a ratio be-
tween variable or total costs di-
videdby total fish production. The
result is a fish market price re-
quired to cover variable or total
costs. Positive (negative) devia-
tion between the breakeven cost
and the actual fish selling price
(U.S. $1.43/kg) will be the profit
(loss) margin. Any breakeven cost
of production covering variable
costs that is less than the market
price for tilapia will show a profit
in the short run. Likewise, when
breakeven costs of production
covering variable and fixed costs
are lower than the selling price,
then long term profitability has
been achieved.
39
RESULTS FROM ENTERPRISE BUDGET ANALYSIS
None of the four Chemical Fertilization treatments
had positive net returns, although treatments N+P/C and
N+P/W had positive income above variable costs (Table
39). Positive income above variable cost indicates a viable
operation in the short run. Changes in the financial climate,
e.g. lower interest rates on operating capital or lowered input
prices or higher fish marketing prices, could transform a
negative net return into a positive value. Thus, enterprises
with short term positive results and negative long-term net
returns should not be completely ruled out for future consid-
eration. The two treatments with tilapia stocked at 10,000/ha
had greater losses than the two treatments where tilapia
stocking rates were 20,000/ha.
Breakeven costs to cover variable and total (vari-
able plus fixed costs) costs for each treatment are shown in
Table 39. Using the N+P/C treatment as an example, the
$1.13 breakeven price to cover variable costs is less than the
1992 Honduran tilapia market price of $1.43, indicating a
positive, short run profit margin of $0.30/kg of fish pro-
duced. Breakeven selling price of tilapia to cover variable
costs for the Chem Fert/Cycle Htreatment was $1.51, greater
than the prevailing tilapia market price ($1.43) and an
example of a negative profit margin in the short run.
Only three of the 13 Organic Fertilization treat-
ments had positive net returns: 20,000/ha + CL, C:N control
and CL 1,000 (Table 39). The treatment with tilapia stocked
at 2,500/ha had the greatest negative net returns, as well as
negative income above variable costs. Organic fertilization
as the sole nutrient input for tilapia production was not
39). A stocking rate of 10,000 tilapia per hectare was
common among these seven treatments. Treatments stocked
at 20,000 tilapia per hectare yielded positive net returns
except for the No aeration/sub. treatment that had positive
income above variable costs but negative net returns. The
economics of nightly or emergency aeration were not deter-
mined during these experiments, but should be investigated
in future research.
Four treatments comprised the Feed Only category
(Table 39). Negative net returns were obtained in the treat-
ment where tilapia were stocked at 10,000/ha, which sug-
gests again a minimum stocking rate of 20,000 tilapia/ha for
profitable fish culture especially where prepared rations are
used. Seasonal effects likely were responsible for differ-
ences in net returns observed for the two treatments stocked
at 20,000 tilapia/ha. The warm-season treatment (20,000/ha
+ feed/W) had positive net returns while the cool-season
treatment (20,000/ha + feed/C) had negative net returns.
Positive net returns also were obtained for the treatment
stocked with 30,000 tilapia/ha, the highest stocking rate
tested. Further research should be conducted on the effect on
yields of seasonal variation and on the use of higher stocking
rates (> 20,000 tilapia/ha), particularly when commercial
fish rations are used.
SUMMARY OF ENTERPRISE BUDGET ANALYSIS
Figure 20 graphically summarizes individual treat-
ments within nutrient input categories that have positive net
returns. The Chemical Fertilization Only category is not
represented because no treatment had positive net returns.
economically feasible under 1992 economic
conditions in Honduras. However, it should
be noted that eight of nine treatments with
negative net returns had positive incomes above
variable costs.
Seventy-eight percent of treatments
in the Organic Plus Chemical Fertilization
category had positive net returns (Table 39).
One treatment with a negative net return, CL
750/Urea, was stocked at 10,000 tilapia/ha,
while all but one of the other treatments had
positive net returns when stocked at 20,000
fish/ha. Both treatments that exhibited nega-
tive net returns had positive income above
variable costs. Temperature effects on pro-
duction between seasons were clearly evident
where treatments were tested both seasons.
Analysis of the first seven Organic
Fertilization Plus Supplemental Feed treat-
ments resulted in negative net returns (Table
30,000/ha + feed/C
20,000/ha + feed/W
30
% 
sat./sub.
10% sat./sub.
CL plus feed
CL 60/feed
CL 250/N + P/W
CL 500/N + P/W
CL 250/N + P/C
CL 500/N + P/C
C4:N1
C6:N1
C8:N1
C:N control
CL 1,000
20,000/ha + CL
|
Feed only
Organic
fertilization
+ feed
Organic +
chemical
fertilization
Organic
fertilization
$0 $500 $1,000 
$1,500 $2,000
Net returns to land and management (US $/ha per 6 months)
Figure 20. Tilapia pond management strategies in Honduras that had positive
net returns to land and management.
40
+n~+nrl
_ ......... .......
Thus, in the Honduran context use of only chemical fertiliz-
ers for production of tilapia is not profitable, at least for
fertilizer quantities tested and interest rates used in this
budget analysis. Some treatments from each of the other four
nutrient-input categories are represented in Figure 20. All
treatments with positive net returns used stocking rates of at
least 20,000 tilapia/ha. This result strongly indicates that
stocking 10,000 tilapia/ha does not fully utilize available
pond food resources and at a higher stocking rate additional
fish yield is possible using the same amount of inputs.
30,000/ha + feed/C
20,000/ha + feed/W
30% sat./sub.
10% sat./sub.
CL plus feed
CL 60/feed
CL 250/N + P/W
CL 500/N + P/W
CL 250/N + P/C
CL 500/N + P/C
C4:N1
C6:N1
C8:N1
C:N control
CL 1,000
20,000/ha + CL
Feed only
Organic
fertilization
+ feed
Organic +
chemical
fertilization
Organic
fertilization
000
-I - -- I
CL..plus. feed 
frlzi
CL.00/ +. PCO
C .N 
chemical.. 
..
0 1,000 2,000 3,000 4.000
Gross fish yield (kg/ha per 5 months)
Figure 21. Comparisons of gross fish yields in relation to different pond
management strategies in Honduras.
30,000/ha + feed/C IFeed 
only
20,000/ha + feed/W
30% sat./sub.
10% sat./sub. 
Organic
CL plus feed 
fertilization
CL 60/plus feed 
+ feed
CL 60/feed I l
CL 250/N + P/W
CL 500/N + P/W
CL 250/N + P/C
CL 500/N + P/C 
Organic +
chemical
C4:N1 ....
fertilization
.. .. .. .. ... ..! . ... ! ........... . ...... ..... .. . ... ...
C 6 :N 1 
-------------------------
C8:N1
C:N control
CL 1,000 Organic
20,000/ha + CL 
fertilization
0 10 20 30 40 50 60 70 80
Return to average investment (%)
Figure 22. Comparison of returns to average investment in relation to different
tilapia pond management strategies in Honduras.
5,
Treatments that utilized organic or chemical fertil-
izers as nutrient inputs had lower production costs than
treatments that used formulated feeds exclusively through-
out the five-month cycle because fertilizers had lower unit
costs. Formulated fish feeds are expensive primarily be-
cause of the cost of protein included in the diet. Competition
for protein feed sources with land livestock activities, the
expense of construction and operation of feed mils, trans-
portation of feed ingredients to the mill and product to
market, packaging, and intermediary-stage value added steps
all contribute to the relatively higher cost of formulated
"~" -~~~~~ "'~~r----- ----- ------ --------~-~
41
-1
feeds. Chemical fertilizers have similar cost
stages but there are no feed spoilage problems
such as those associated with formulated feeds.
However, chemical fertilizer prices often are
subsidized by governments as a stimulus to
production agriculture. Manures as a source of
pond nutrients probably have the least add-on
expenses included in its final price or cost (real
or opportunity). Labor to collect, bag and
transport are the main items determining the
prices for chicken litter and cow manure used
in these experiments.
The use of formulated feeds, stocking
rates of 20,000 and 30,000 fish/ha of pond
surface area and aeration increased fish yield
( Figure 21). However, it is likely that critical
standing crop or carrying capacity of ponds
were not attained with these management
practices because of the low stocking rates.
Stocking rates in the range of 30,000 to
60,000 fish/ha of pond surface area need to be
investigated.
The production system with the greatest
amount of production is not always the most
profitable enterprise. Comparing the summary
of positive net returns in Figure 20 with the
summary of production yields in Figure 21, it
can be seen that production from treatment
C6:N1 was lower than six other treatments
(CL60/feed, CL plus feed, 10 percent sat./sub,
30 percent sat./sub, 20,000/ha + feed/W and
30,000/ha +feed) but net returns from C6:N1
was greater than all other treatments. Con-
versely, the 30,000/ha +feed/C treatment had
the greatest production yield (Figure 21), but
only resulted in intermediate net returns
when compared to other treatments (Figure
20). It is the cost of intensified inputs, such as
additional fingerlings at higher stocking
rates, the greater quantity and quality of
nutrients required, aeration
equipment and associated
electricity costs, etc., that
decreased the greater sales
revenue amount to an over-
all intermediate net return.
Thus, variable costs and
new equipment required
will determine if greater fish
production results in higher
profits to the enterprise.
Season affected
fish yields, which in turn
caused changes in profit-
ability. In the Feed Only
category two similar experi-
ments (20,000/ha + feed)
run in different seasons,
warm/rainy and cool/dry,
resulted in positive and
negative netreturns, respec-
tively. The three greatest
net returns recorded, for
treatments C6:N1, CL Plus
Feed and CL60/Feed, oc-
curred during the warm/
rainy season. However, sea-
sonal effects were not al-
ways the sole factor to in-
fluence profitability; sea-
son by treatment interac-
tions also appeared to af-
fect profitability. For ex-
ample, the cool/dry-season
N+P/C treatment resulted
in high losses, while the
CL500/N+P/C treatment
produced positive net re-
turns. However, CL500/
N+P/W (warm/rainy sea-
son) produced net returns
TABLE 40. EFFECT OF VARYING COST OF 45.4-KG SACK OF COMMERCIAL FISH FEED ON
SHORT-TERM (INCOME ABOVE VARIABLE COSTS) AND LONG-TERM (NET RETURNS TO
LAND AND MANAGEMENT) PROFITABILITY OF AQUACULTURE ENTERPRISES IN HONDURAS.
VALUES ARE IN U.S. DOLLARS PER SIX-MONTH PRODUCTION CYCLE
Commercial fish feed price (per 45.4 kg)
Income above variable costs Net returns to land and management
Treatment -20 pct. 1992 price + 20 pct. -20 pct. 1992 price +20 pct.
($9.63) ($12.04) ($14.44) ($9.63) ($12.04) ($14.44)
CL 60/feed 3,190 2,799 2,407 1,804 1,412 1,020
20,000/ha + feed/W 2,797 2,260 1,723 1,519 982 446
CL plus feed 3,124 2,898 2,673 1,736 1,511 1,285
CL 500/0.5% feed 370 338 305 -1,017 -1,050 -1,082
CL 500/1.0% feed 564 494 424 -824 -893 -963
CL 500/2.0% feed 443 290 137 -944 -1,097 -1,250
CL 750/1 mon. sub. 722 403 83 -665 -984 -1,304
CL 750/2 mon. sub. 575 307 39 -812 -1,080 -1,349
CL 750/3 mon. sub. 884 694 503 -503 -693 -884
10,000/ha + feed/C 1,099 836 572 -179 -442 -705
20,000/ha + feed/C 1,574 1,024 474 297 -254 -804
30,000/ha + feed/C 2,629 2,015 1,400 1,352 737 123
No aeration/sub. 1,543 1,146 749 156 -241 -638
10% sat./sub. 2,531,, 2,134 1,737 1,058 660 263
30% sat./sub. 2,683 2,286 1,889 1,210 812 415
TABLE 41. EFFECT OF VARYING COST OF 50-KG SACK OF UREA FERTILIZER ON
SHORT-TERM (INCOME ABOVE VARIABLE COSTS) AND LONG-TERM (NET RETURNS TO
LAND AND MANAGEMENT) PROFITABILITY OF AQUACULTURE ENTERPRISES IN HONDURAS.
VALUES ARE IN U.S. DOLLARS PER SIX-MONTH PRODUCTION CYCLE
Urea fertilizer price (per 50 kg)
Income above variable costs Net returns to land and management
Treatment -20 pct. 1992 price +20 pct. -20 pct. 1992 price +20 pct.
($10.07) ($12.60) ($15.11) ($10.07) ($12.60) ($15.11)
Chem Fert/Cycle II -60 -102 -144 -1,337 -1,379 -1,422
CL 750/urea 827 799 771 -450 -478 -507
C8:N1 1,927 1,918 1,910 649 641 632
C6:N1 3,191 3,171 3,151 1,913 1,893 1,874
C4:N1 1,719 1,677 1,635 442 400 357
CL 750/N+P/C 654 612 570 -624 -666 -708
CL 500/N+P/C 1,516 1,465 1,415 238 188 137
CL 250/N+P/C 1,419 1,362 1,306 141 85 29
N+P/C 618 562 506 -659 -715 -771
CL 500/N+P/W 3,071 3,026 2,981 1,794 1,749 1,704
CL 250/N+P/W 1,353 1,303 1,252 76 25 -25
N+P/W 958 904 851 -320 -373 -427
almost ten times higher than CL500/N+P/C. Treatment
CL250/N+P/C (cool/dry season) had higher net returns than
treatment CL250/N+P/W (warm/rainy season). Effect of
seasonal variation on fish yield in Honduras also should be
investigated more thoroughly.
Percentage return on investment (ROI) values for
treatments having positive net returns ranged from one
percent to 77 percent (Figure 22). ROI values are important
because potential investors use these values for comparison
to alternative investment opportunities.
SENSITIVITY ANALYSES
Sensitivity analyses were conducted for the major
production inputs, adult tilapia sales price and variable
capital interest rate. Rather than use historical price pat-
terns, which were unavailable, to indicate upper and lower
limits for sensitivity analyses, price ranges were estimated
through interviews with Honduran and expatriate aquacul-
tural officials. Ranges then were increased slightly to be
more encompassing. Sensitivity analyses were used to deter-
mine how specific price changes affected short- and long-
term profitability of 41 enterprises (Table 36).
42
TABLE 42. EFFECT OF VARYING SALES PRICE OF MARKETABLE TILAPIA ON
SHORT-TERM (INCOME ABOVE VARIABLE COSTS) AND LONG-TERM (NET RETURNS TO
LAND AND MANAGEMENT) PROFITABILITY OF AQUACULTURE ENTERPRISES IN HONDURAS.
VALUES ARE IN U.S. DOLLARS PER SIX-MONTH PRODUCTION CYCLE
Marketable tilapia sales price (per kg)
Income above variable costs Net returns to land and management
Treatment -25 pct. 1992 price +25 pct. -25 pct. 1992 price + 25 pct.
($1.07) ($1.43) ($1.80) ($1.07) ($1.43) ($1.80)
Chem Fert/Cycle I
Chem Fert/Cycle II
Cow Manure
CL 125
CL 250
CL 500
CL 1,000
CL 1,000 @ 2
2,500/ha + CL
10,000/ha + CL
20,000/ha + CL
CL 750/no urea
CL 750/urea
CL 60/feed
20,000/ha + feed/W
CL plus feed
CL 500/no feed
CL 500/0.5% feed
CL 500/1.0% feed
CL 500/2.0% feed
CL 750/no sub.
CL 750/1 mon. sub.
CL 750/2 mon. sub.
CL 750/3 mon. sub.
10,000/ha + feed/C
20,000/ha + feed/C
30,000/ha + feed/C
No aeration/sub.
10% sat./sub.
30% sat./sub.
C:N control
C8:N1
C6:N1
C4:N1
CL 750/N+P/C
CL 500/N+P/C
CL 250/N+P/C
N+P/C
CL 500/N+P/W
CL 250/N+P/W
N+P/W
-837
-563
493
-88
272
490
680
299
-590
380
789
124
162
1,191
670
1,468
-250
-163
-93
-350
244
-431
-469
-95
-21
-295
302
-64
665
768
1,051
946
1,876
723
-71
603
529
-92
1,760
479
177
-680
-102
1,071
325
819
1,168
1,508
1,083
-307
1,058
1,677
704
799
2,799
2,260
2,898
169
338
494
290
876
403
307
694
836
1,024
2,015
1,146
2,134
2,286
2,041
1,918
3,171
1,677
612
1,465
1,362
562
3,026
1,303
904
-523
359
1,649
738
1,365
1,844
2,336
1,866
-24
1,736
2,565
1,284
1,436
4,407
3,849
4,328
589
839
1,080
929
1,508
1,236
1,084
1,483
1,692
2,342
3,727
2,356
3,604
3,804
3,030
2,890
4,465
2,631
1,295
2,328
2,196
1,216
4,292
2,126
1,631
-2,114
-1,841
-784
-1,366
-1,005
-787
-598
-978
-1,868
-898
-488
-1,153
-1,115
-196
-607
81
-1,527
-1,551
-1,480
-1,737
-1,033
-1,818
-1,857
-1,482
-1,299
-1,572
-975
-1,452
-809
-705
-227
-331
599
-554
-1,349
-675
-749
-1,369
483
-799
-1,100
-1,957
-1,379
-206
-953
-459
-110
230
-195
-1,584
-220
399
-573
-478
1,412
982
1,511
-1,108
-1,050
-893
-1,097
-401
-984
-1,080
-693
-442
-254
737
-241
660
812
763
641
1,893
400
-666
188
85
-715
1,749
25
-373
-1,800
-918
372
-539
88
-566
1,059
589
-1,301
459
1,287
7
158
3,020
2,572
2,941
-688
-549
-307
-458
231
-151
-304
96
415
1,065
2,450
969
2,130
2,330
1,753
1,613
3,188
1,353
17
1,050
919
-61
3,014
849
354
most sensitive to changes
in feed price. At stocking
rates of 20,000/ha, enter-
prises were less sensitive
to feed price. A stocking
rate of 30,000/ha was prof-
itable at all feed prices
evaluated. A +/- 20 percent
change in feed price re-
sulted in, on average, a 33
percent change in short-
term profitability and a 56
percent change in long-term
profitability.
All enterprises, except
one, that used urea were
profitable in the short term
in response to a +/- 20 per-
cent change in urea price
(Table 41). Urea price
variation tested resulted in
an average change in short-
term profitability of 7 per-
cent and an average change
in long-termprofitability of
29 percent. A +/- 10 
per-
cent change in
diammonium phosphate
price caused, on average, a
9 percent change in short-
term enterprise profitabil-
ity. Long-term profitabil-
ity changed 12 percent, on
average, in response to price
variation. Further, long-
term profitability was nega-
tive where DAP was used
with urea, but was positive
when DAP was combined
with chicken litter.
Tilapia production inputs evaluated by sensitivity
analysis were: commercial fish feed (20 to 25 percent
protein), urea, diammonium phosphate (DAP), adult tilapia
sales price and variable capital interest rate.
Change in commercial feed price of +/- 20 percent
did not reverse the sign for income above variable cost for
any enterprise that used feed (Table 40). Reduction of feed
price by 20 percent resulted in transition of negative net
returns into positive returns for two enterprises (Table 40).
Treatments where tilapia were stocked at 10,000/ha were
43
Variation in tilapia
sales price by +/- 25 percent of the 1992 level resulted in
negative short-term profitability (Table 42). Twenty-five
treatments had positive income above variable costs at all
tilapia sales prices. Twelve treatments had negative income
above variable costs at the lowest tilapia sales price. A 25
percent-reduction in tilapia sales price resulted in negative
long-term profitability for most treatments (Table 42).
Changes in short- and long-term profitability averaged 96
percent and 200 percent, respectively, 
in response to the +/
- 25 percent change in tilapia sales price.
Variable capital
interest rates have been
volatile in Honduras in re-
cent years, and have varied
by as much as 50 percent.
Fifty percent increase or
decrease in interest rate was
tested by sensitivity analy-
sis (Table 43). All treat-
ments except three showed
positive short-term profit-
ability at the three interest
rates tested. Long-term
profitability was negative
for 25 treatments, which
indicates that short-term
profitability was marginal.
Fourteen treatments had
positive long-term profit-
ability at all interest rates
tested. The +/- 50 percent
variation in variable capital
interest rate yielded aver-
age changes in short- and
long-termprofitability of 17
percent and 37 percent, re-
spectively.
ACKNOWLEDGEMENTS
The authors thank
the entire professional and sup-
port staff, past and present, of
the El Carao National Fish
Culture Research Center for
their untiring support of this
project since 1983. We also
thanks officials at the
Chem Fert/Cycle I
Chem Fert/Cycle II
Cow Manure
CL 125
CL 250
CL 500
CL 1,000
CL 1,000 @ 2
2,500/ha + CL
10,000/ha + CL
20,000/ha + CL
CL 750/no urea
CL 750/urea
CL60/feed
20,000/ha + feed/W
CL plus feed
CL 500/no feed
CL 500/0.5% feed
CL 500/1.0% feed
CL 500/2.0% feed
CL 750/no sub.
CL 750/1 mon. sub.
CL 750/2 mon. sub.
CL 750/3 mon. sub.
10,000/ha + feed/C
20,000/ha + feed/C
30,000/ha + feed/C
No aeration/sub.
10% sat./sub.
30% sat./sub.
C:N control
C8:N1
C6:N1
C4:N1
CL 750/N+P/C
CL 500/N+P/C
CL 250/N+P/C
N+P/C
CL 500/N/P/W
CL 250/N+P/W
N+P/W
-612
-1
1,136
394
890
1,248
1,603
1,189
-230
1,147
1,778
789
891
2,989
2,477
3,047
248
424
590
408
966
558
455
825
970
1,244
2,266
1,339
2,330
2,484
2,141
2,021
3,277
1,789
722
1,569
1,465
669
3,133
1,407
1,009
Direcci6n General de Pesca y Acuicultura and the Administration
Department, Ministry of Natural Resources, for their support
throughout this project. Students and faculty of the Biology Depart-
ment, Universidad Nacional Aut6noma de Honduras are thanked
for their collaboration during this project. This research was funded
by U.S. Agency for International Development (USAID) as part of
the Pond Dynamics/Aquaculture Collaborative Research Support
Program (grants: DAN-4023-G-SS-2074-00, DAN-4023-G-SS-
7066-00 and DAN-4023-G-00-0031-00), USAID/Honduras as part
of contract 522-0168-C-00-8010-00, Auburn University and the
Secretarfa de Recursos Naturales, Republic of Honduras.
LITERATURE CITED
APHA. 1989. Standard methods for the examination of water and
-680
-102
1,071
325
819
1,167
1,508
1,083
-307
1,058
1,677
704
799
2,799
2,260
2,898
169
338
494
290
876
403
307
694
836
1,024
2,015
1,146
2,134
2,286
2,041
1,918
3,171
1,677
612
1,465
1,362
562
3,026
1,303
904
-747
-203
1,007
256
747
1,086
1,413
976
-384
969
1,575
619
707
2,609
2,043
2,749
91
251
398
172
787
247
159
563
701
803
1,763
953
1,938
2,088
1,940
1,815
3,065
1,565
501
1,361
1,259
455
2,919
1,199
799
-1,890
-1,278
-141
-883
-387
-30
325
-88
-1,507
-131
501
-488
-386
1,602
1,199
1,659
-1,030
-963
-797
-980
-312
-829
-932
-562
-307
-33
988
-48
856
1,011
864
744
1,999
511
-555
292
188
-608
1,855
129
-269
-1,957
-1,379
-206
-953
-459
-110
230
-195
-1,584
-220
399
-573
-478
1,412
982
1,511
-1,108
-1,050
-893
-1,097
-401
-984
-1,080
-693
-442
-254
737
-241
660
812
763
641
1,893
400
-666
188
85
-715
1,749
25
-373
-2,025
-1,481
-271
-1,022
-531
-191
136
-301
-1,661
-308
298
-658
-571
1,221
765
1,362
-1,186
-1,136
-990
-1,215
-490
-1,140
-1,228
-824
-576
-474
486
-435
465
614
662
538
1,788
288
-766
84
-18
-822
1,642
-79
-478
waste water, 17th Edition. American Public Health Associa-
tion, Washington, D.C.
Boyd, C. E. 1979. Water quality in warmwater fish ponds.
Auburn University Agricultural Experiment Station, Au-
burn University, AL 36849 U.S.A.
Boyd, C. E., and D. R. Teichert-Coddington. 1992. Relationship
between wind speed and reaeration in small aquaculture
ponds. Aquacultural Engineering 11: 121-131.
Egna, H. S., N. Brown, and M. Leslie (editors). 1987. Pond
Dynamics/Aquaculture Collaborative Research Support Pro-
gram Data report: Volume one: General Reference: Site
descriptions, materials and methods for the global experi-
ment. Oregon State University, Corvallis, OR.
Guerrero, R. D., and W. L. Shelton. 1974. An aceto-carmine
squash method for sexing juvenile fish. The Progressive
Fish-Culturist 36:56.
44
TABLE 43. EFFECT OF VARYING ANNUAL INTEREST RATE (APR) OF VARIABLE CAPITAL
ON SHORT-TERM (INCOME ABOVE VARIABLE COSTS) AND LONG-TERM (NET RETURNS
TO LAND AND MANAGEMENT) PROFITABILITY OF AQUACULTURE ENTERPRISES IN HONDURAS.
VALUES ARE IN U.S. DOLLARS PER SIX-MONTH PRODUCTION CYCLE
Variable capital interest rate (APR)
Income above variable costs Net returns to land and management
Treatment -50 pct. 1992 rate +50 pct. -50 pct. 1992 rate +50 pet.
(11.5 pct.) (23 pct.) (34.5 pct.) (11.5 pct.) (23 pct.) (34.5 pct.)
I
7 "fC 'Y'renrrier
JdU
Economic conclusions lrow iw uaeeorise age
analysis of 41 a rrmr, n t-iput regimes hs tee tohe veniica
(ion of a important agricultural economic concep and ao
some specific insighits regarding Honduras. Aprinciple well
kniown ta agricultural ecoromists, Ihe corcapt of marginal
rewoaps, is wel demonstrated w-hh the Honduras data. Sim
oply stated, the aooacultral production system with 'the
goravest production, is not always the enterprise producing
-ie greatest 
proAit
Tnc 23 rarent hicees 
rate fr gcI
rena-Able 'or main on ce croi !bile nu-
irianil Thoul reies a -npre lnbam now
Temporal financia conditiors hav man 'npartarn
'Impact a. ale cnoi-e of 5sh 
productior systems5; the 
higher
ne interest rate IIe less fish farm aperators are likely fo
borrow money for aparaing capital or for capital exendi-
auras. Rural rasa mce-lmilad farmers may never borrow
monay aid onsad-aty 0 t-adito nil agorcu ural patterns a1
supply nutrieats s. 2., aires, an ' 
vvocte 2 e.i ,
family labor.
Sfckhng r m a at e-s 2DU3 U~hh pane
surrae area appe aa . aar to qidTo reT
fabe aisb culaira colaprse -*n Hc oduras.
Oily ooe enterprise where , ap!2a were stockedea
1 0,000/ha had posK (Va reet returns, o land, l abor and manaoye-
ment, ailthouigh many treatm'ent saC this sir ckirg rate showed
roositive income above var able costs.
Eena it fhclarc also appear '
2n'li Of Tl ood Mragrera
rcerms yeiiera
1 
y wre g~e2
varm/ram season.i
'a"_,ec prati
ss'ters; mc
ar dorirg the
Honralsc?L/ , 23RS rzeascaae sh ve pr-,
seame ra~u soma ction op aonsi te ado-
ra Msahfa rer anil d avrsesaiofonmtr'ent -
ajpiC my' acs Las amule ia praftalL
2002 iltur,9 orem" 0 59s even ci tha 23 oar-
cent acubrT loan nlaresl Tae prevail-
irg 'r ednd as in -992. Twe ma tram ren's
rr'n -four ccyeori a of ma ma il alirnge-
mal nt tedu Ir in pan r ma. - alretrms, 
T
bese
mmri2 romnpgae a s'esrs poiiO de a9
cen- o iro le cpt lo. s a Tisb farmacts in
cens '!ivtc 'nd dc'a'ed inal emta';rise
poroa Lilit wa / s e Mfecia.
1 
by Cha2nges in
Mam~ sals Prce ozd rie Mad b1,e raxt
Aeaenmie
1 
n c ar and long-termn em-
erre i rolibln, followed byv ariable
caperi tnMc] Oz rate. 'aparad "haileresl
rate chngeO~s tic rat nal'ee' Irelr at plfn-
251i ty as mue, as was m'lcipaiad " Ia
Changes rin chmel ica hlr prices had ii""e
lece-' a.' enlerprise pratalili y.
On ie ow technaology end of Chis range, usa af
2 i cka'l 'iter resulte in profita ble tilacia production. Onihe
nighiect seioy end of this 'ange, use of formula ted feed ace.
atnr rasuite 4n profitable tilaria production. anth
idd'cle of 'C echnologym rge, tha conmbinationis of chocker
'tera ad chemical 
2
'Irti ira or or chicken litte" and formi-
l.aed fae resultad i rofitabhle production systems. Such a
-an ge e~-f
1
h produetion sys ems a laws appro-
priate a> ahoices' ear areas as well as an conditions
favorirg cap taiinterstication. A wida choice of profitable
er.odacton iatensitie sand system alternatives a far efficient
resource' utiliationand leads to afounda~ iorsustaiable
aquaco aure.
PROJECT PUBLICATIONS AND PRESENTATIONS
Green, B. W., and D.R. Teichert-Coddington. 1994. Growth of
control and androgen-treated Nile tilapia during treatment,
nursery and grow-out phases in tropical fish ponds. Aquac-
ulture and Fisheries Management 25: In press.
Green, B. W., and C. E. Boyd. 1994. Chemical budgets in
organically fertilized fish ponds in the dry tropics. World
Aquaculture '94 conference, New Orleans, LA, 12 to 18
January 1994.
Green, B. W., and C. E. Boyd. 1994. Water budgets for fish ponds
in the dry tropics. World Aquaculture '94 conference, New
Orleans, LA, 12 to 18 January 1994.
Ayub, M., C. E. Boyd and D. R. Teichert-Coddington. 1993.
Effedcts of urea application, aeration and drying on total
carbon concentrations in pond bottom soils. The Progres-
sive Fish-Culturist 55: 210-213.
Green, B. W., and D.R. Teichert-Coddington. 1993. Production of
Oreochromis niloticus fry for hormonal sex reversal in
relation to temperature. Journal of Applied Ichthyology
9:230-236.
Teichert-Coddington, D. R. 1993. Development of production
technologies for semi-intensive fish farming during the past
decade in Central America. Symposium on Aquacultural
Research in Central America, San Jose, Costa Rica, 25 to 29
October 1993.
Teichert-Coddington, D. R., and B.W. Green. 1993. Usefulness of
inorganic nitrogen in organically fertilized tilapia produc-
tion ponds. World Aquaculture '93 conference,
Torremolinos, Spain, 26 to 28 May 1993.
Teichert-Coddington, D. R., and B. W. Green. 1993. Tilapia yield
improvement through maintenance of minimal dissolved
oxygen concentrations in experimental grow-out ponds in
Honduras. Aquaculture 118: 63-71.
Teichert-Coddington, D. R., and B. W. Green. 1993. Comparison
of two techniques for determining community respiration in
tropical fish ponds. Aquaculture 114: 41-50.
Teichert-Coddington, D.R., and B. W. Green. 1993. Influence of
daylight and incubation interval on dark-bottle respiration in
tropical fish ponds. Hydrobiologia 250: 159-165.
Teichert-Coddington, D. R., B. W. Green, C. Boyd, R. Gomez and
N. Claros. 1993. Substitution of inorganic nitrogen and
phosphorus for chicken litter in production of tilapia. In H.
S. Egna, M. McNamara, J. Bowman and N. Astin (editors).
Tenth annual administrative report. Pond Dynamics/Aquac-
ulture Collaborative Research Support Program. Oregon
State University, Corvallis, OR.
Teichert-Coddington, D. R., B. W. Green, M. I. Rodriguez, R.
Gomez and L. A. Lopez. 1993. On-farm testing of PD/A
CRSP fish production systems in Honduras. In H. S. Egna,
M. McNamara, J. Bowman and N. Astin (editors). Tenth
annual administrative report. Pond Dynamics/Aquaculture
Collaborative Research Support Program. Oregon State
University, Corvallis, OR.
Green, B. W. 1992. Substitution of organic manure for pelleted
feed in tilapia production. Aquaculture, 101: 213-222.
Green, B. W., D. R. Teichert-Coddington, and L. A. Lopez. 1992.
Production of Oreochromis niloticus fry in earthen ponds
for hormonal sex inversion. Twenty-third Annual World
Aquaculture Conference, May 21 to 25, 1992, Orlando,
Florida.
Teichert-Coddington, D. R., R. Gomez, M. Ponce and H. Ramos.
1992. Reproduction of guapote tigre in earthen ponds:
female to male stocking ratios. In H. S. Egna, M. McNamara
and N. Widner (editors). Ninth annual administrative re-
port, Pond Dynamics/Aquaculture Collaborative Research
SupportProgram, 1991. Oregon State University, Corvallis,
OR.-
Teichert-Coddington, D. R., and B. W. Green. 1992 Yield
improvement through maintenance of minimal oxygen con-
centrations in tilapia grow-out ponds in Honduras. Twenty-
third Annual World Aquaculture Conference, May 21 to 25,
1992, Orlando, Florida.
Teichert-Coddington, D.R., B. W. Green and R.P. Phelps. 1992.
Influence of water quality, season and site on tilapiaproduc-
tion in Panama and Honduras. Aquaculture, 105: 297-314.
Teichert-Coddington, D. R., B. W. Green, C. E. Boyd and M. I.
Rodriguez. 1992. Supplemental nitrogen fertilization of
organically fertilized ponds: Variation of the C:N ratio.In H.
S. Egna, M. McNamara and N. Weidner (editors). Ninth
annual administrative report. Pond Dynamics/Aquaculture
Collaborative Research Support Program. Oregon State
University, Corvallis, OR.
Boyd, C. E., D. R. Teichert-Coddington and B. W. Green. 1991.
Change of fish pond soils during culture and drying periods.
In H..S. Egna, J. Bowman and M. McNamara (editors).
Eighth annual administrativereport. PondDynamics/Aquac-
ulture Collaborative Research Support Program. Oregon
State University, Corvallis, OR.
Green, B. W., and D.R. Teichert-Coddington. 1991. Growth of
normal and sex reversed Oreochromis niloticus during hor-
mone treatment, nursery and grow-out phases in earthen
ponds. Twenty-second Annual World Aquaculture Confer-
ence, June 16 to 20, 1991, San Juan, Puerto Rico.
Green, B. W., and D.R. Teichert-Coddington. 1991. A comparison
of two samplers used with an automated data acquisition
system in whole-pond community metabolism studies. The
Progressive Fish-Culturist, 53(4): 236-242.
Green, B. W., and D. R. Teichert-Coddington. 1991. Effect of fry
stocking rate, hormone treatment duration and temperature
on the production of sex-reversed Oreochromis niloticus. In
H. S. Egna, J. Bowman and M. McNamara (editors). Eighth
annual administrative report. Pond Dynamics/Aquaculture
Collaborative Research Support Program. Oregon State
University, Corvallis, OR.
Green, B. W., and D. R. Teichert-Coddington. 1991. Production
of Oreochromis niloticus fry in earthen ponds for hormonal
sex reversal. In H. S. Egna, J. Bowman and M. McNamara
(editors). Eighth annual administrative report. Pond Dy-
namics/Aquaculture Collaborative Research Support Pro-
gram. Oregon State University, Corvallis, OR.
Teichert-Coddington, D.R., and B. W. Green. 1991. Determina-
tion of nighttime oxygen respiration in fish culture ponds in
Honduras: Effect of measurement methodology. Twenty-
second Annual World Aquaculture Conference, June 16 to
20, 1991, San Juan, Puerto Rico.
Teichert-Coddington, D. R., B. W. Green and C. E. Boyd. 1991.
Benthic respiration in newly renovated ponds. In H. S.
Egna, J. Bowman and M. McNamara (editors). Eighth
annual administrative report. Pond Dynamics/Aquaculture
Collaborative Research Support Program. Oregon State
University, Corvallis, OR.
46
Teichert-Coddington, D. R., B. W. Green and M. I. Rodriguez.
1991. Relative influence of feed and organic fertilization on
polyculture of tambaquf and tilapia. In H. S. Egna, J.
Bowman and M. McNamara (editors). Eighth annual ad-
ministrative report. Pond Dynamics/Aquaculture Collabo-
rative Research Support Program. Oregon State University,
Corvallis, OR.
Teichert-Coddington, D. R., B. W. Green, C. E. Boyd and M. I.
Rodriguez. 1991. Supplemental nitrogen fertilization of
organically fertilized ponds. In H. S. Egna, J. Bowman and
M. McNamara (editors). Eighth annual administrative re-
port. Pond Dynamics/Aquaculture Collaborative Research
Support Program. Oregon State University, Corvallis, OR.
Teichert-Coddington, D. R., B. W. Green, N. Claros and M. I.
Rodriguez. 1991. Optimization of feeding in combination
with organic fertilization. In H. S. Egna, J. Bowman and M.
McNamara (editors). Eighth annual administrative report.
Pond Dynamics/Aquaculture Collaborative Research Sup-
port Program. Oregon State University, Corvallis, OR.
Green, B. W. 1990. Substitution of organic manure for pelleted
feed in tilapia production. EIFAC/FAO Symposium on
Production Enhancement in Still Water Pond Culture. Prague,
Czechoslovakia.
Green, B.W., H.R. Alvarenga, R.P. Phelps and J. Espinoza. 1990.
Pond Dynamics/Aquaculture Collaborative Research Data
Reports: Honduras Project, Cycle l of the global experiment.
Vol. 6, No. 1. Oregon State University, Corvallis, OR. 94
pp.
Green, B.W., H.R. Alvarenga, R.P. Phelps and J. Espinoza. 1990.
Pond Dynamics/Aquaculture Collaborative Research Data
Reports: Honduras Project, Cycle II of the global experi-
ment. Vol. 6, No. 2. Oregon State University, Corvallis,
OR. 94 pp.
Green, B. W., and L. A. L6pez. 1990. Factabilidad de la producci6n
masiva de alevines machos de Tilapia nilotica a trav6s de la
inversi6n hormonal de sexo en Honduras. Agronomfa
Mesoamericana, 1:21-25.
Green, B. W., and D.R. Teichert-Coddington. 1990. Comparison
of two sampler designs for use with automated data acqui-
sition systems in whole-pond community metabolism stud-
ies. EIFAC/FAO Symposium on Production Enhancement
in Still Water Pond Culture. Prague, Czechoslovakia.
Green, B. W., D.R. Teichert-Coddington and R.P. Phelps. 1990.
Response of tilapia production and economics to varying
rates of organic fertilization in two tropical American coun-
tries. Aquaculture, 90: 279-290.
Popma, T. J. and B. W. Green. 1990. Aquaculture production
manual: Sex reversal of tilapiain earthen ponds. Res. and
Dev. Series 35, International Center for Aquaculture, Au-
burn Univ., AL.
Teichert-Coddington, D.R., and B. W. Green. 1990. Influence of
primary productivity, season and site on tilapia production
in organically fertilized ponds in two Central American
countries. EIFAC/FAO Symposium on Production En-
hancement in Still Water Pond Culture. Prague, Czechoslo-
vakia.
Alvarenga, H.R. and B.W. Green. 1989. Producci6n y aspectos
econ6micos del cultivo de tilapia en estanques fertilizados
con gallinaza. Revista Latinoamericana de Acuicultura,
40:35-39.
Green, B.W. and H.R. Alvarenga. 1989. Efecto de diferentes tazas
de gallinaza en la producci6n de tilapia. Revista
Latinoamericana de Acuicultura, 40:31-34.
Green, B.W. and H.R. Alvarenga. 1989. Tilapia production
systems based on organic fertilization and feed. Thirty-fifth
Annual Meeting, Programa Cooperativo Centroamericano
para el Mejoramiento de Cultivos Alimenticios, San Pedro
Sula, Honduras. (in Spanish)
Green, B.W., H.R. Alvarenga, R.P. Phelps and J. Espinoza. 1989.
Pond Dynamics/Aquaculture Collaborative Research Data
Reports: Honduras Project, Cycle III of the global experi-
ment. Vol. 6, No. 3. Oregon State University, Corvallis,
OR. 114 pp.
Green, B.W. and L.A. Lopez. 1989. Feasibility of mass producing
hormonally sex-reversed Tilapia nilotica fingerlings in
Honduras. Thirty-fifth Annual Meeting, Programa
Cooperativo Centroamericano para el Mejoramiento de
Cultivos Alimenticios, San Pedro Sula, Honduras. (in Span-
ish)
Green, B.W., R.P. Phelps and H.R. Alvarenga. 1989. The effect
of manures and chemical fertilizers on the production of
Oreochromis niloticus in earthen ponds. Aquaculture, 76:37-
42.
Teichert-Coddington, D., B. Green, and M. I. Rodriguez. 1989.
Effect of feeding rate on tilapia production in ponds fertil-
ized with chicken litter. Thirty-fifth Annual Meeting,
Programa Cooperativo Centroamericano para el
Mejoramiento de Cultivos Alimenticios, San Pedro Sula,
Honduras. (in Spanish)
Alvarenga, H.R., B.W. Green and R.P. Phelps. 1988. Production
and economic aspects of tilapia culture in ponds fertilized
with chicken litter. Thirty-fourth Annual Meeting, Programa
Cooperativo Centroamericano para el Mejoramiento de
Cultivos Alimenticios, San Jose, Costa Rica. (in Spanish)
Green, B. W. 1988. Honduras freshwater aquaculture project: final
technical report. Unpublished report, Auburn University,
Alabama. 56 pp.
Green, B.W., H.R. Alvarenga and R.P. Phelps. 1988. The effect
of stocking rate on the production of Tilapia nilotica in
ponds. Thirty-fourth Annual Meeting, Programa Cooperativo
Centroamericano para el Mejoramiento de Cultivos
Alimenticios, San Jose, Costa Rica. (in Spanish)
Green, B.W. and H.R. Alvarenga. 1987. Intensive fingerling
production of hybrid tilapia (Tilapia nilotica x Tilapia
hornorum). Eighteenth Annual Meeting, World Aquacul-
ture Society, Guayaquil, Ecuador.
Green, B.W. and H.R. Alvarenga. 1987. Effect of chicken litter
fertilization rate on the production of tilapia. Presented at
33
rd Annual Meeting, Programa Centroamericano para 
el
Mejoramiento de Cultivos Alimenticios, Guatemala City,
Guatemala. (in Spanish)
Green, B.W., R.P. Phelps and H.R. Alvarenga. 1987. The effect
of nitrogen and phosphorus sources in fertilizers used for the
production of Tilapia nilotica. Eighteenth Annual Meeting,
World Aquaculture Society, Guayaquil, Ecuador.
Alvarenga, H.R. and B.W. Green. 1986. Production and growth of
male Tilapia nilotica and of hybrid Tilapia nilotica x Tilapia
hornorum in ponds. Rev. Latinoamer.de Acuicultura, 29:6-
10. (in Spanish)
47
APPENDIX
Cross-Reference of Studies to PD/A CRSP Work Plans
Study number in this report PD/A CRSP work plan
Study Al Work plan 2: site-specific study
Study A2 Work plan 5: Study 3
Study A3 Work plan 5: Study 4
Study A4 Work plan 5: site-specific study
Study B I Work plan 1
Study B2 Work plan 2
Study B3 Work plan 3
Study B4 AU-USAID/Honduras project
Study B5 Work plan
Study B6 Work plan 5: Study 6
Study B7 Work plan 6: Study 1
Study Cl AU-USAID/Honduras project
Study C2 AU-USAID/Honduras project
Study C3 Work plan 4 (1987-1988)
Study C4 Work plan 5: Study 1
Study Dl Work plan 4 (1988-1989)
Study D2 Work plan 5: Study 2
Study D3 Work plan 5: Study 5
Study D4 Work plan 5
Study El Work plan 1: site-specific study
Study E2 Work plan 2: site-specific study
Study E3 Work plan 4 (1988-1989)
Study E4 Work plan 5: Study 5

<a a4
:4 "a ' "a a~aa'~4 ~ <a a ~ 4
> a a~aV~ <aa~~a>~<"a'a'a a aaa~~
faij;~
44<'fr>4" > a a>
a>
a a a aa
'a
a ~a a a
a 4
a, " 4 a
a >~
a a'
>4 a
a <a a
a ,~a' 4
a a a a'
> !'a>
a a> a a
aa a'a '~
'a a
a a
a"
aa a
''a a a
'a a
aa <a <a
4 a
a >, 'aa
a aa
a> a
a a 4 <a' 4a a a
a , a
a a
<a a a a>> a
a' a'
4 'a a~ a' a a a a
a a a
a~ a', aa a 'a a> a
a, a a a
4> 4~
<a a <a '>'a <a
~aaaa - a' 'a a -aa>~aaaa<~a~~
- /aa - 'a -~ :4 -- aaaaaaa.
aaa a a a a a' ~4<j ~4 a'~ a
a 'a
>aa'a:4~a4a4~; ~a
a a 'a "aa">< <a> a 444 ,aa' a
a a a 'a aa a ' a'a''~
4
a<~'aa'aa<'a'a~4,:4a>~>,a,
'a a a 4 'a
a aa a'a a 'a a a >~<a'aaaaaaa'a aa4<a~a'a'
a a>a'4a a <a - - -- ~ -- -
a aa a
4 .a- ~aaaa4
a' a'a ' a ,a 'a
a' a a a 4' a.
a'~> a ~ a~ a -- a a' < '<a ' > 4~'>~
- >~- 4aa~aa >~aaaa
a '4< a a< a a ,a
aa.aa>, a 'a
a a 'a4'a4'aa.'aa>4~aaa'~
~-<a>~ 4, a 'a a 'a"" - a'> a,>
>4 'a' '4 'a> a ' " a a
- 'a> a 4
,a'a~aaaa,>-aa'a a>,a.<a
aa<aa'aa4aa,$aa'a4a'4aaaaaaaa4aaa4la ''a < 'a a a4a' a a a a'a'a
~ - a, -
4
a> aaa44aaa 4 a
,a'aa<"4a<'4,aa a'
- " 'a
'aaaa'I'~'a~'}4>4 >a' a' a>' < a" 'a a ,'a. -"<'>4" ;~-~d'~aia
7 ~
aaaaaaa a'a' a ~
a a' a':> a>:>' 4&h<~~f'~a a .a
a> a a a a a, ,4a4~>~>4a>'j4>fa4'<> aa'
4
' a>~~~aa >4< a->''a< -
a" a 4'aa'a"a--'
a a
a a~:4~aa:,>;~ a-
>, a'a~aa >aa> a a - > a a'a" aa a' a~aaaaaa a a> ~--~
aaaaaaaa,4aa<~4 a <,'''''> <''
-a,' 'a ~4a.>4~aa'~a<~<a<a' a / aa"aa.a>-><a~>
a' <a '>4'a~ a'
a a': -a>a.aaa.
a,>
- - a. - - aaaaa'a'aa'a a'a<a aA',>aaaa'a'a~a'a 
><a,<~- 4 4':' a'>