Tilapia Fry and Fingerling Production

 in Small Tanks

 James E. Rakocy and Ayyappan Nair
 CVI Agricultural Experiment Station
 P.O. Box 920, Kingshill
 St. Croix, U.S. Virgin Islands 00850


 Two experiments were conducted to evaluate the use of
small tanks as a hatchery for the production of tilapia fry and
fingerlings. The effect of stocking density of brood fish on fry
production was evaluated using a sex ratio of four females to
one male. Brood fish (Tilapia aurea) were stocked at rates of
2.7, 5.4 and 8.1 fish/m2 in nylon spawning nets (hapas)
suspended in tanks. Fry were collected seven times at 2-week
intervals. Mean fry production was 12.2, 15.1 and
23.8/m2/hapa/day or 5.3, 3.9 and 3.8/female/day, respec-
tively. The fry were stocked at rates of 26, 52, 78, 104, 130 and
155/m2 to determine the optimum rate for fingerling produc-
tion. The fry were fed four times daily for 63 days. The daily


feeding rate was 15 % of initial body weight for three weeks
followed by weekly adjustments to 5%. Predation by dragonf-
ly larvae caused wide variation in survival (23-88%). Fry at the
26/m2 rate grew fastest, averaging 2.3 g with 37% survival.
13.5 % actual feeding rate and 2.3 feed conversion ratio. Fry at
the 104/m2 rate weighed 1.2 g with 88% survival, 6% actual
feeding rate and 0.75 feed conversion ratio. The data indicates
that the daily feeding rate should be higher than 5% of body
weight for maximum growth but should not exceed 40
kg/ha/day in small static tanks.
Keywords: tilapia, hatchery systems, aquaculture.


 In many areas, tilapia producers do not have a commercial
source of fingerlings for growout and must therefore produce
their own. Their operations may be too small to justify the con-
struction of brood and nursery ponds or the necessary resources
may not be available. Small tanks may be the the appropriate
unit for fry and fingerling production to satisfy their needs.
 Several investigators have used tanks in studies involving
tilapia fry or fingerling production. Uchida and King (1962) ex-
amined the production and growth of Tilapia mossambica fry in
tanks. Shelton et al. (1978) used tanks for producing T. area fry
for sex reversal experiments. Snow et al. (1983) studied methods
of removing T. area from tanks and raising them to fingerling
size. Hughes and Behrends (1983) suspended net enclosures in
tanks to study the mass production of T. nilotica seed (eggs, sac
fry and advanced fry).
 The use of net enclosures (hapas) in tanks is a new practice that
facilitates fry production. Swim-up fry form schools at the surface
and stay close to the tank walls for a few days. At this stage they
are easily caught with a dip net and transferred to a nursery unit.
However, fry that avoid capture prey upon subsequent spawns
and production declines (Uchida and King, 1962; Snow et al.,
1983). With hapas, this problem is eliminated as all the fry can
be frequently removed by crowding them to one end of the hapa
and capturing them with a dip net after the brood fish have been
gently removed by hand to the cleared portion of the hapa.
 An important factor in the production of fry is the density of
brood fish (Hughes and Behrends, 1983). The growth rate of fry
to fingerling size is also dependent on density, among other fac-
tors (Uchida and King, 1962; and Snow et al., 1983). The objec-
tives of this study were to evaluate the effects of brood fish stock-
ing density on fry production and the stocking level of fry on their
growth rates to fingerling size (5g).

MATERIALS AND METHODS
 This study was conducted at the College of the Virgin Islands
 Agricultural Experiment Station in St. Croix, United States
 Virgin Islands.


236


Fry Production
 Fry were produced in nylon hapas suspended in circular tanks.
The hapas consisted of 1.6-mm mesh and were 3.05 m long, 1.22
m wide and 1.22 m deep with a surface area of 3.72m2. The
vinyl-lined tanks were 3.65 m in diameter and 1.22 m deep with
a surface area of 10.51 m2. The tanks were maintained at a water
depth of 1 m and were occasionally aerated when fish exhibited
signs of oxygen stress. During the study, water temperature rang-
ed from 22 to 26C.
 Brood fish (Tilapia area) were stocked in the hapas at three
densities (Table 1) with each treatment density replicated twice.
The desired sex ratio for each treatment was four females to one
male, but five errors in sex identification occurred as a total of 120
brood fish were stocked, greatly altering the sex ratio among
treatments.
 The brood fish were stocked on 25 November 1981 and fed at a
rate of 1% of the initial body weight per day. The feeding rate of
the high density treatment was reduced to 0.5% of the initial
body weight from day 52 through 91 and increased to 0.75%
from day 92 through 117. The experiment was terminated on 22
March 1982.
 Fry were collected initially on day 34 and thereafter at 14-day
intervals for a total of seven collections. Only advanced, free
swimming fry were counted.

Fingerling Production
 As fry were collected and counted, they were stocked in circular
tanks to determine the effect of density on growth. The tanks
were vinyl-lined steel swimming pools with a surface area of
9.65 m. The water level was maintained at a depth of 0.8 m. Fry
were stocked at densities of 26, 52, 78, 104, 130 and 155 fish/m2.
Each treatment density was replicated three times. Since there
were insufficient numbers of fry to stock all the tanks at one time,
fry were stocked as they became available during collection from
spawning hapas. Fry were stocked four times over a 6-week period
(Table 2). The fry were fed Purina Trout Chow (#1), four times
daily for 63 consecutive days beginning the day after stocking.

 PROCEEDINGS of the CARIBBEAN FOOD CROPS SOCIETY-VOL. XX