personatus in Aquaria
by Todd Gardner
One of the first problems often encountered in the first
larval rearing attempts of marine fish species, is finding an
appropriate first food. Although rotifers, particularly those
of the genus, Brachionus are probably the most important
and widely used first food in marine fish culture today (Hoff
and Snell, 1993), many species will not eat them (Young, 1995).
Rotifers often constitute a major portion of the zooplankton
in estuaries. Therefore it is likely that many fish species
whose larval stages develop in estuaries are adapted for utilizing
them as a first food (Hoff and Snell, 1993). Stomach content
analyses of oceanic species such as groupers however, have revealed
a larval diet dominated by copepods (Grover et al.,1998). Young
(1993) states, after more than 20 years of rearing marine fish,
that there are two major factors influencing prey selection
in marine fish larvae: prey size relative to mouth size, and
prey movement. Our inability to provide live foods of appropriate
size and exhibiting an appropriate swimming motion to elicit
a feeding response, may be a major impediment to the advancement
The purpose of these experiments was to investigate the
Carribean goby, Coryphpterus personatus, as a candidate
for commercial production at a marine ornamental fish hatchery
in Puerto Rico. This paper describes the development of a technique
for feeding and rearing the larvae of C. personatus which
was found in early trials, not to accept rotifers as a first
Materials and Methods
Five adult specimens of Coryphopterus personatus
were placed in a 20 gallon aquarium with three 2-inch lengths
of 1-inch PVC pipe. The aquarium was part of a 5000 gallon broodstock
system consisting of 200 20 and 40 gallon aquaria. Central filtration
was achieved with two Magnum‘ 1/2hp Jacuzzi pumps and a 400
gallon sump. One pump circulated water between the sump and
aquaria with approximately 20% flow being bypassed through a
foam fractionator (protein skimmer). The other pump circulated
water between the sump and the remaining filter elements which
included, in series, a Hayward‘ rapid sand pool filter, an ultraviolet
sterilizer (six bulbs, 48" each), and an 800 gallon container
filled with plastic bio balls. The broodstock system was maintained
at ambient temperature (23-30 C) in a semi-enclosed greenhouse.
Broodstock fish were fed four times per day with a gelatin-based
diet containing a variety of fresh seafood and vegetables, kelp
meal, powdered Spirulina, lecithin, pelleted salmon food
and a multi-vitamin formulated for marine fish.
16 days after the C. personatus were introduced
into the system, a spawn of approximately 300 eggs was found
in one of the PVC pipes. Although the eggs were not accurately
counted or measured, initial inspection revealed that they are
considerably smaller than eggs of Gobiosoma species.
This first spawn was left in the broodstock tank and sacrificed
in order to determine incubation time. On the morning of the
sixth day, the eggs were gone and a few Larvae were observed
in the aquarium. Three larvae were removed and measured (approximately)
under a dissecting scope, at 2 mm.
Three days after the first hatch, another spawn was found
and estimated to be about 20% larger than the first spawn. On
the afternoon of the fifth day after spawning, the PVC pipe
containing the eggs was removed and placed in a 300 gallon,
round black polyethylene larval rearing tank filled with sterilized
seawater and pre-inoculated with Isochrysis galbana and six
million rotifers (Brachionus plicatilis). A ceramic air
stone with moderate flow was placed halfway down the center
standpipe. The PVC pipe was suspended vertically, one inch below
the surface and a second air stone with light air flow was positioned
just below it, so that the entire interior of the pipe was brushed
with a slow, steady stream of bubbles. The larval rearing tank
was part of a 30,000 gallon system with filtration identical
to that of the broodstock system except that the bio ball filter
was replaced with an 800 gallon fluidized bed filter utilizing
beach sand as the filter medium. Tanks in trials 1-6 of this
experiment were never connected to the system. The only water
exchange in these tanks was in the form of daily additions of
approximately 20 gallons of moderately dense Isochrysis in seawater.
Pelagic microalgae acts as a natural water conditioner and food
for planktonic larval foods such as rotifers. No attempt was
made to count algae cells. The system was outdoors and covered
with 85% shadecloth.
The next morning, the spawn was found to have hatched completely
and larvae were observed drifting around the tank, evenly dispersed.
By the afternoon of day 2, post-hatch, the larvae were exhibiting
negative phototaxis, congregating in the shadow of the tank
wall. At this time, they also began to exhibit hunting behavior,
swimming up to rotifers and other particles in the water column
and occasionally striking at them; however no successful strikes
were observed and microscopic examination of three larvae revealed
no food in their digestive tracts. By day 5, post-hatch, the
number of larvae appeared to be in decline and by day 7, no
larvae could be found in the tank.
For the remainder of these rearing trials, a spawn of approximately
300-500 eggs was encountered in the adult tank every 7-10 days.
No further attention will be given to spawning in this paper.
Trials 2-6 were treated exactly as trial 1 except that rotifers
were replaced with a variety of unidentified ciliate species,
ranging in size from 25m to 80 m. The ciliates were obtained
from contaminated rotifer cultures by passing culture water
through 2 consecutive Nitex‘ sieves of 80m and 25m , respectively.
This procedure was repeated each morning. Ciliate concentrations
in the larval rearing tanks were counted twice each day under
a dissecting microscope, and averaged 20 organisms/ml, one hour
after feeding, and 3/ml immediately before the next feeding,
24 hours later.
In trials 2, 3, 4 and 6 no larvae were seen after day 7
In trial 5 larvae were observed eating after day 3. At
day 10 the number of larvae appeared to be significantly reduced
and no growth was apparent. On day 12, only 3 larvae were seen.
The last larva was spotted in the beam of a flashlight on the
night of day 15.
At the beginning of trial 7, a contaminating dinoflagellate
caused the Isochrysis cultures to crash. As a result,
an algae substitute was developed in order to sustain zooplankton
populations in larval tanks. The preparation consisted of 3g
of powdered Spirulina, 1g of Protein Selco‘ and 2 cups
of water, blended in an electric mixer for 1 minute. For trial
7 this preparation was used daily in place of Isochrysis.
Ciliates were added on day 1, using the technique described
above. On day 2 no new ciliates were added as the concentration
in the rearing tank was found to be greater than 20/ml. The
ciliate concentration remained high (greater than 10/ml) through
day 20. Additionally, rotifers appeared in the tank and were
at a concentration of 12/ml by day 20, although no rotifers
were intentionally introduced.
On day 20 larvae were observed feeding on rotifers, although
it is likely that they had already been eating rotifers for
several days. On day 21 water in the larval rearing tank was
tested for ammonia using an Aquarium Systems‘ ammonia test kit,
however an accurate measurement could not be achieved as the
ammonia concentration in the tank exceeded the range of the
kit. The tank was immediately connected to the larval rearing
system and the water supply was opened sufficiently to acheive
an exchange rate of approximately one gallon per hour (a moderate
drip). The water flow was gradually increased over the next
five days until a rate of approximately 40 gallons per hour
Beginning on day 22, 5-10 million rotifers (depending on
availability) were added, daily to replace those lost to water
exchange and increased fish predation.
At day 40, a large number of fish larvae were still seen
in the tank. One fish was removed and measured at 8mm. A small
number of one-day-old Artemia nauplii were introduced
into the tank and were rapidly consumed. Beginning on day 40
three feedings of approximately 250,000 Artemia nauplii
would be administered each day. All Artemia were enriched
for at least 12 hours with Protein Selco‘ prior to feeding.
Rotifer additions to the larval rearing tank were discontinued
after day 45.
On day 58, several fish were found to have taken on the
pale, silvery-orange, characteristic of adult C. personatus.
Approximately 90% of the fish had undergone the change at day
65, and by day 68, metamorphosis appeared to be complete. Between
days 58 and 70, 14 dead larvae and juveniles were found and
removed from the tank. These were the only mortalities observed
in trial 7. Gradually, over the next 3 weeks the water flow
rate was increased to 200 Gallons per hour and the juvenile
gobies were weaned onto a diet of commercial dry food.
Results and Discussion
Ten months after trial 7 began, 544 adult C.personatus
were counted, moved into a new tank and sold on the wholesale
market for $7.50 each. The procedure described for trial 7 was
repeated three more times with C. personatus and once
with C. dicrus, with nearly identical results achieved
each time. The procedure was also repeated numerous times with
the small larvae of Ptereleotris zebra , Liopropoma
eukrines and Gobiodon citrinus with zero survival
after day 10, each time.
Since no serious effort was made to identify the many species
of zooplankton present in the larval rearing tanks in these
experiments, it is impossible to say what the larvae were feeding
on prior to accepting rotifers. Although the reasons for larval
survival in trial 7 remain unknown, here are three likely possibilities
: 1. The algae substitute stimulated a bloom of some ciliate
species that were an acceptable food for larval C. personatus
and were not available in sufficient numbers in the presence
of natural algae; 2. the algae substitute, fed on by ciliates
acted to enhance the nutritional value of the ciliates, making
them a healthier diet; and 3. the larval fish ingested the algae
substitute, directly. Our inability to successfully rear the
larvae of P. zebra ,L. eukrines and G. citrinus
using this technique, may have resulted from a lower tolerance
for elevated environmental ammonia concentrations in these species,
or from specific dietary requirements which as yet, have not
been met. These are questions to be answered in future investigations.
Grover, Jill J., David B. Eggleston and Jonathan M. Shenker.
1998. Transition from pelagic to demersal phase in early juvenile
nassau grouper, Epinephelus striatus: pigmentation, squamation
and ontogeny of diet. Bulletin of Marine Science, 62(1):79-113.
Hoff, Frank H. and Terry W. Snell. 1993. Plankton Culture
Manual. Florida Aqua Farms, Inc.
Young, Forrest A. 1995. Experimental rearing of wild collected
fish eggs. FAMA, 18(2):178-179.
Young, Forrest A. 1993. Live foods for marine fish larvae.