1 1 2 1,3 4

1 1

1 Department of Fisheries and Aquatic Sciences, Moi University, Eldoret, Kenya; 2 Department of Wildlife, Oregon State

University, Corvallis, OR, USA; 3 Department of Aquatic Ecology and Ecotoxicology, Amsterdam, The Netherlands;4 Department of Biological Sciences, Moi University, Eldoret, Kenya

Problems of limited number of dry feeds as supplement or

replacement of live feeds have led to poor larval nutrition in

many species of fish. Therefore, the suitability of co-feeding

8-day-old African catfish (Clarias gariepinus) posthatch lar-

vae using live feed (Artemia salina) and formulated dry diet

containing freshwater atyid shrimp (Caridina nilotica) during

weaning was investigated. The experiment ended after

21 days of culture and respective groups compared on the

basis of growth performance, survival, feed utilization and

nutrient utilization. Larvae co-fed using 50% Artemia and

50% formulated dry diet resulted in significantly (P < 0.05)

better growth performance, food gain ratio (FGR), protein

efficiency ratio (PER) and productive protein values (PPV)

than other treatments. The lowest growth performance

occurred in larvae weaned using 100% formulated and

commercial dry diets. Better survival of over 90% was

obtained in larvae weaned using 50% Artemia and 50% dry

diet, while abrupt weaning using 100% dry diets resulted in

lower survival (<75%). These results support a recommen-

dation of co-feeding C. gariepinus larvae using a formulated

dry diet containing C. nilotica and 50% live feed when

weaning is performed after 8 days posthatching period.

KEY WORDSKEY WORDS: Artemia nauplii, Clarias gariepinus larvae,

formulated feed, growth, nutrient utilization

Received 18 May 2009, accepted 28 September 2009

Correspondence: Victoria Chepkirui-Boit, Department of Fisheries and

Aquatic Sciences, Moi University, P.O. Box 1125, Eldoret, Kenya. E-mail:

vcboit@yahoo.com

African catfish (Clarias gariepinus, Burchell 1822) is one of

the most important fish species currently being cultured both

within and outside its natural range of tropical and sub-

tropical environments (Adewolu et al. 2008). Its resistance to

diseases, high fecundity and easy larvicidal production in

captivity makes it of commercial importance (Hogendoorn

1979; Haylor 1991; Kestemont et al. 2007). However, there

are still considerable challenges in larval nutrition of this fish.

During early stages of growth, the larvae fish rely on the

yolk sac for nutritional requirements (Kolkvoski 2001). At

the onset of exogenous feeding, live feeds such as brine

shrimp (Artemia nauplii), yeast, zooplankton and unicellular

algae are more appropriate because the fish have difficulty

assimilating dry prepared diets (Hogendoorn 1979; Verreth &

Den Bieman 1987; Verreth & Van Tongeren 1989; Verreth

1994). However, during weaning several dry diets are intro-

duced (Verreth et al. 1987, 1993; Verreth & Van Tongeren

1989; Awaiss & Kestemont 1998) . Absolute dependency on

live feeds as the dietary source is a major constraint in larval

nutritional research. Most of these dry diets are still mainly

based on dried decapsulated cysts of Artemia (Verreth et al.

1987; Verreth & Den Bieman 1987; Kerdchuen & Legendre

1994) or on alkan yeast (Hecht & Appelbaum 1987).

The continued utilization of Artemia as live and dry diets is

likely to bring considerable challenges because of the inten-

sive production techniques as well as the cost of Artemia

production. To reduce reliance on Artemia feeds, research

has been conducted to provide alternative sources of dry

diets (Gonzalez et al. 2008). These dry diets must, however,

meet the nutritional requirement and should be readily

2011 17; e82–e89. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

doi: 10.1111/j.1365-2095.2009.00737.x

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

� 2010 Blackwell Publishing Ltd

Aquaculture Nutrition

accepted by the larval fish. Freshwater atyid shrimp (Cari-

dina nilotica Roux) is promising as a test ingredient for fish

feeds (Liti et al. 2006; Rasowo et al. 2008). Under practical

conditions, growth of Nile tilapia (Oreochromis niloticus) was

comparable to other fish meal diets when C. nilotica was the

key protein ingredient (Liti et al. 2006). The feedstuff has

good palatability and high nutritional quality (of up to 66%

crude protein, 10%, crude lipid, 6% carbohydrates) and

offers opportunities of formulating a micro-diet applicable in

weaning C. gariepinus larvae. For this, the present study

evaluated the effects of feeding live feeds, the Artemia nauplii

and a dry micro-diet formulated with C. nilotica on feed

utilization, growth, survival and nutrient utilization during

weaning of C. gariepinus.

The 21-day experiment was carried out under controlled

conditions at the Moi University, Eldoret Kenya in the

Department of Aquaculture and Fisheries Science. Three

mature female broodstock (Mean weight = 280 ± 5.5 g)

and two mature males (mean weight = 300 ± 8.1 g) were

seined from the central broodstock ponds at the institution�s

research farm and transferred to the hatchery. Fertilization

and incubation of the eggs followed protocols detailed in de

Graaf et al. (1995). Water temperature in the incubator was

maintained at 27 ± 0.1 �C using a thermostat heater. After

hatching, 500 larvae each (mean weight = 3.5 ± 0.2 mg)

were siphoned out of the incubation tank and transferred

into 15 glass aquaria of 20 l water volume each in a recir-

culating water system. The facilities were aerated by means of

electric pump with air stones throughout the 21 days study

period. Three extra aquaria were not stocked with larvae to

help in quantifying larval ingestions. The water flow through

each aquarium was maintained at a 0.4 L min)1 to ensure

a renewal rate of at least once every 1 h. The tap water was

stocked in an intermediate tank for at least 48 h to remove

chlorine before using it in the rearing facilities. To ensure that

bacteria are destroyed in all aquaria, the tap water was

passed through germicide UV lamp. Water temperature in

the aquaria was maintained at 27.0 ± 0.5 �C using thermo-

stat heaters. A natural photoperiod of 12 h/12 h light/dark

photoperiod was maintained. Dissolved oxygen (DO) was

measured each morning (JENWAY 3405 electrochemical

analyser (Barloword Scientific Ltd, Essex, UK) with values

ranging from 6–8 mg L)1. Ammonia and nitrites were mea-

sured from water samples taken in the tanks weekly; values

of mean were always <0.02 mg L)1 and nitrites <0.05 mg

L)1. The third day after hatching, the larvae were transferred

to the aquaria and fed Artemia only. Our experiment started

on the same day of the transfer.

Formulation of the experimental diet was made using

C. nilotica as an ingredient to include high level protein in the

diet. The C. nilotica was oven-dried at 30 �C for 6 h before

being ground using an electric meat grinder (Model: SM-G70;

Guangzhou Sunmile Industries, Guangzhou, China).

To formulate the experimental diets, the ingredients were

mixed in proportions provided in Table 1. Dietary ingredi-

ents were ground and passed through a 0.05-mm mesh sieve

and homogenized for 3 min in a blender (Hobart M-600;

Hobart Corp.,Troy, OH, USA). Mustard oil and premixes

was gradually added, and warm water (approximately 50%

of the total weight) was added during mixing. Simon-Heese

pelleting machine (Boxtel, The Netherlands) was used to

pelletize the wet mixture after addition of cassava as binders.

The 300-lm pellets obtained were dried in a forced-air oven

Table 1 Proximate composition (g kg)1) of the formulated diet

Ingredients

Formulated

diet

Caridina nilotica1 755

Wheat flour 185

Mustard oil2 15

Binders (Cassava) 15

Vitamin premix and Mineral premix3 15

Salt (NaCl) 15

Proximate composition

Dry matter 901

Crude protein 553

Crude lipid 93

Ash 98

Crude fibre 64

Nitrogen-free extracts (NFE) 93

Gross energy (MJ kg)1)4 17.8

1 Obtained locally from Lake Victoria.2 Imported from USA.3 Commercial formula (mg premix kg)1 diet). Union de Empresas

de Piensos MINAGRI, Cuba. Vitamins: retinol, 12 500 000 IU; thia-

mine, 10 000 mg; riboflavin, 20 000 mg; pyridoxine, 1000 mg;

cyanocobalamine, 40 mg; ascorbic acid (Stay C), 500 000 mg; cho-

lecalciferol, 2 400 000 IU; a tocopherol, 100 000 mg; pantothenic

acid, 40 000 mg; choline chloride, 1 600 000 mg; folic acid,

2000 mg; nicotinic acid, 140 000 mg; biotin, 1000 mg; inositol,

300 000 mg; paraminobenzoic acid, 35 000 mg. Minerals (mg):

cobalt, 200; copper, 2000; iron, 20 000; iodine, 1500; manganese,

40 000; zinc, 20 000; selenium, 100.4 Gross energy was calculated on the basis of 23.0 kJ CP, 38.1 kJ

Crude lipid and 17.2 kJ Carbohydrate (Tacon 1990).

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

at 45 �C for 4 h. The pellets were packed in plastic bags and

refrigerated at 4 �C until use.

Artemia nauplii were obtained by hatching 2.5 g of Artemia

salina cysts (from Great Salt Lake, UT, USA) in a cone

sedimentation chamber with 35 g common salt in 1 L of

dechlorinated tap water. Temperature was maintained at

28 �C by use of a thermostat inserted into the chamber. They

were cultured in 5 l plastic containers with 4 l of 17 g L)1

brackish water (prepared by mixing 2 l of distilled water and

2 l of natural seawater, filtered in 47-mm glass microfibre

filter) with strong aeration, at a density of 10 ind. mL)1.

After 1 day, newly hatched Artemia nauplii were collected

using a 100-lm sieve and distributed equally to two 2 l plastic

containers half-filled with 17 g L)1 artificial water, provided

with strong aeration and placed in a water bath maintained

at 25 ± 1.0 �C.

Ingredients, diet and carcass were analysed for proximate

composition to determine the exact quantity needed for the

formulation of the diet. Dry matter (DM) was determined by

oven drying the ingredients at 110 �C for 24 h. Crude protein

(N · 6.25) was determined by Kjelhdahl method after acid

digestion. Ash content was determined by incineration in a

muffle furnace at 550 �C for 24 h. Crude fibre was deter-

mined by digestion with 1.25% H2SO4 and 1.25% NaOH

solutions. Nitrogen-free extracts (NFE) were calculated from

the differences. Gross energy was calculated using conversion

factors for protein, lipids and carbohydrates provided in

Tacon (1990).

Five types of diets (D1: 100% Artemia, D2: 50% Art-

emia + 50% formulated diet, D3: 100% formulated diet,

D4: 100% commercial catfish diet, D5: 50% Artemia + 50%

commercial catfish diet) were used for the experiment.

Commercial catfish diet (D4) was used for comparison of the

results against the formulated diet.

Each feeding experiment was conducted in triplicate.

Before and during feeding, water inflow was interrupted for

at least 10 min before normal circulation was re-established.

Preweighed portions of each of the formulated diet were

taken and hand fed to the fish (4% body weight) four times a

day between 08:00 h and 22:00 h. The Artemia nauplii were

added 4–6 per day. Feeding levels on Artemia corresponded

to a near satiation level and was calculated according to the

procedure of Verreth & Den Bieman (1987). To feed the fish

to satiation and minimize the remaining, feed in each tank

was counted for each aquarium and the amount of feed fed

was adjusted accordingly.

Sampling was performed during stocking in aquaria (day 3),

during weaning (day 8) and after every 3 days till last day.

Ten fish from each experimental tank were removed (without

replacement) and individually weighed daily to monitor

growth. Feed supply was then adjusted based on the

remaining larvae biomass in the aquaria.

Estimation of Artemia biomass ingested was based on the

number of Artemia per litre, and biomass of individual

Artemia estimated by using phytoplankton analyser (Model

Phyto-EDF Heinzwalz GmbH D-91090, Effeltrich,

Germany). Technique described by Kamler et al. (1986) was

Table 2 Parameters of growth and nutrient utilization in Clarias gariepinus larvae after 21 days of feeding

Dietary treatments*

D1 D2 D3 D4 D5

Initial mean weight (mg) at day 8 13.6 ± 0.2 13.2 ± 0.4 13.9 ± 0.2 13.1 ± 0.2 14.6 ± 0.3

Final mean weight (mg) at day 21 64.2 ± 3.2b 84.2 ± 3.9d 55.2 ± 2.8a 74.1 ± 2.5c 52.3 ± 2.7a

Mean weight gain (mg) 50. ± 2.6b 71.0 ± 3.2d 41.3 ± 2.1a 59.9 ± 2.1c 39.2 ± 2.0a

SGR (% day)1) 11.8 ± 0.6b 14.3 ± 0.7d 10.6 ± 0.5a 12.7 ± 0.6c 10.6 ± 0.4a

% Survival 95.3 ± 2.9c 92.3 ± 2.1c 65.9 ± 4.9a 94.4 ± 3.1c 74.3 ± 2.1b

FCR 1.1 ± 0.2b 0.6 ± 0.2a 1.6 ± 0.3c 0.8 ± 0.2a 1.5 ± 0.3c

PER 1.2 ± 0.3a 2.7 ± 0.3b 1.3 ± 0.4a 2.5 ± 0.3b 1.4 ± 0.3a

PPV 13.3 ± 1.7a 24.2 ± 5.1b 10.8 ± 4.9a 23.2 ± 4.0b 9.9 ± 5.1a

* Mean within a row having different superscript are significantly different (P < 0.05).

D1, 100% Artemia nauplii; D2, 50% Artemia nauplii + 50% formulated diet; D3, 100% formulated diet; D4, 50% Artemia nauplii + 50%

catfish starter feed; D5, 100% catfish starter feed. SGR, specific growth rate; FCR, food conversion ratio; PER, protein efficiency ratio; PPV,

productive protein value.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

used to calculate the daily feed consumption of the Artemia

per larvae. This method measures the changes in the number

of prey. Larval ingestion rates was calculated as:

I ¼ 24 P n�1;

where I is ingestion rate; P is the predation (number of

Artemia consumed by larvae in 1 h); n = number of larvae

per aquarium. The predation of larvae was further estimated

by considering the changes of larvae density in the control

group as an exponential function:

dNc

dt¼ bNc orNc24 ¼ N0:eb:24

Simplifying this equation:

b ¼ 1

24ln

Nc24

N0;

where Nc = initial density of the Artemia in the control

group (without larvae); Nc24 = final density of the Artemia

in the control group after 24 h; N0 = initial density of Art-

emia per aquarium; b = Artemia hatching rate. The changes

in Artemia density in the aquarium containing larvae were

described by the equation:

dNf

dt¼ bNf � P ¼ NfðtÞ ¼ ðN0 �

PbÞebt þ P

b

Simplification of the equation yields:

P ¼ bðNc24 � Nf24Þðeb

24 � 1Þ;

where Nf 24 = final density of Artemia in the experimental

group containing the larvae after 24 h exposure to larval

predation.

Growth in weight of the fish was expressed as the specific

growth rate (SGR, % day)1) using the formula SGR (%

day)1) = (eg)1)100, where g = (ln(W2))(ln(W1))(t2)t1))1

and W2 and W1 are weights on day t2 and t1, respectively.

Mortalities were determined at each sampling date by

counting the number of dead fish during sampling and per

cent survival calculated based on the number of larvae

remaining in the tank as a percentage of the stocked larvae

(deducting the number of larvae that were sampled).

After determination of growth and survival, the 10 fish

were freeze-dried and finally ground for proximate analysis.

At the start and end of the experiment, individual weights

were measured on an analytical balance (accu-

racy = 0.01 mg) after blotting the larvae individually using

tissue paper according to the standardized procedure. Dry

matter was determined after drying overnight at 110 �C.Nutrient utilization was determined using three parameters:

feed conversion ratio (FCR), protein efficiency ratio (PER)

and protein productive value (PPV, %). Daily feed ration

was calculated from the average individual wet weight

determined by weighing two groups of five fishes and their

dry matter content estimated from a relationship between wet

weight and dry weight detailed in Verreth & Den Bieman

(1987). Calculation of food conversion ratio (FCR) based on

measurements of the dry matter content at the start and at

the end of the experiment was defined as the total food ration

(dry matter) per unit of dry fish weight and was calculated

following protocols of Verreth & Den Bieman (1987). Other

parameters of nutrient utilization were determined as:

PER ¼ ðFB� IBÞWprot�1f

PPV ¼ 100 ðWprot2 �Wprot1ÞWprot�1;

where FB and IB = final and initial larval biomass (g);

Wprot1 and Wprot2 are initial and final protein weight in

larvae, respectively (g); Wprotf = weight of dietary protein

supply per larvae.

The effects of dietary treatments on growth performance and

survival, feed utilization and nutrient utilization were anal-

ysed using one-way ANOVAANOVA after verifying the homogeneity of

variance using �Hartley�s test�. When significant differences

were discerned, treatment means were compared using post

hoc Tukey�s HSD test. All statistical analyses were preformed

using GenStat (GenStat Release 4.2 Discovery Edition, VSN

International Ltd, Lawes Agricultural Trust (Rothamstad

Experimental Station, United Kingdom)). Arcsine transfor-

mation was performed on survival data to conform to the

rules of normality before subjecting it to ANOVAANOVA test. In all

the above analysis, significance was accepted at P < 0.05.

Changes in Artemia ingestion rates are as shown in Fig. 1.

Throughout the rearing period after weaning (after 8 days

posthatching), ingestion of Artemia nauplii remained signi-

ficantly higher in treatment D1. After 4 days of weaning

(after 12 days posthatching), ingestion of the Artemia nauplii

by C. gariepinus larvae was similar in treatment D2 and D4;

however, after 7 days of weaning (15 days posthatching) and

thereafter, higher Artemia nauplii ingestion was discerned in

treatment D2 when compared to D4. Up to 20% reduction in

the Artemia nauplii ingestion was achieved 14 days after

weaning in treatment D2 when compared to D4.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

Growth performance and nutrient utilization parameters

of C. gariepinus larvae in different dietary treatments after

21 days are shown in Table 2. Growth in terms of mean

weight gain and SGR was significantly (P < 0.05) better in

treatment D2 followed by D4. However, treatment D3 and

D5 produced poor mean weight gain and SGR. Treatment

D2 maintained significantly better growth performance than

other dietary treatments 7 days after weaning (Fig. 2). The

mean growth in dietary treatments involving co-feeding

Artemia with commercial dry diet differentiated (P < 0.05)

from control treatment 14 days after weaning. Weaning

using 100% formulated dry diet or commercial dry diet

without live feeds (D3 and D5, respectively) produced low

larval growth performance than the control (D1) throughout

the 14 days postweaning period.

Survival of C. gariepinus larvae was significantly

(P < 0.05) affected by dietary treatments. Higher survival of

over 90% was observed in treatments receiving live feeds (D1)

as well as in treatments receiving 50% dry diets in combina-

tion with live feeds (D2 and D4). In treatments receiving only

dry feeds, lower survival of upto 65% and 75% was reported

in dietary treatments receiving 100% formulated dry diet

(D3) and 100% commercial dry diet (D5), respectively.

Parameters of nutrient utilization efficiencies exhibited

positive relationships with growth of larvae under various

dietary treatments. Treatments D2 and D4 had significantly

(P < 0.05) the lowest FCR followed by control treatment

(D1), which were higher than FCR in treatments D3 and D5.

Higher PER and PPV were reported in dietary treatments

combining live and dry feeds (D2 and D4) than other treat-

ments.

It should be borne in mind that reduction in Artemia is

advantageous because of their high costs as well as com-

plexity in their culture. Therefore, in larval rearing of any

species, diets that reduce dependence on live prey production

are of technical and economic interest, which has led to

intense research towards Artemia replacement diets

(Gonzalez et al. 2008). In this study, weaning to 50%

formulated or commercial dry diets in combination with 50%

Artemia nauplii reduced ingestion of live feed item after

2 weeks of feeding experiment. The reduced ingestion on

Artemia nauplii by C. gariepinus larvae and possible switch-

ing of feeding to the dry diet enabled the larvae to avoid

restricted feeding on Artemia nauplii but combine the

advantage of live and dry diets as reported by Awaiss &

Kestemont (1998). During feeding times, we observed

C. gariepinus larvae pursuing rotifers when only Artemia

was fed, with reduced swimming activity when Artemia was

provided in combination with dry feeds. When feeding was

0

10

20

30

40

50

60

70

80

90

0Time (day)

Mea

n w

eigh

t (m

g)

D1 D2 D3 D4 D5

3 6 9 12 15 18 21

Figure 2 Growth curves for Clarias gariepinus on various feed

treatments during the 21 days study period. Vertical bars denote

standard error of the mean calculated from the mean-square for

error of the ANOVAANOVA. D1 = 100% Artemia nauplii; D2 = 50%

Artemia nauplii + 50% formulated diet; D3 = 100% formulated

diet; D4 = 50% Artemia nauplii + 50% catfish starter feed;

D5 = 100% catfish starter feed.

0

2000

4000

6000

8000

10 000

12 000

14 000

16 000

0Time in days

D1 D2 D4In

gest

ion

rate

s (A

rtem

ia n

aupl

ii–1 d

ay–1

)

3 8 12 15 18 21

Figure 1 Estimated Artemia ingested by Clarias gariepinus larvae at

different days during their growth periods. Vertical bars denote

standard error of the mean calculated from the mean-square

for error of the ANOVAANOVA. D1 = 100% Artemia nauplii; D2 = 50%

Artemia nauplii + 50% formulated diet; D4 = 50% Artemia

nauplii + 50% catfish starter feed.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

carried out with only Artemia as the main feed item, higher

Artemia consumption was reported than co-feeding. This

indicates that C. gariepinus larvae exploited Artemia for their

nutritional demand, albeit at higher supply of the Artemia.

At similar feeding levels involving combination of dry diets,

lower Artemia ingestion was established when formulated

diet containing C. nilotica was used when compared to the

commercial catfish feed. It is postulated that supply of the

formulated diet with C. nilotica ingredient encouraged larvae

to rely on formulated feed at the expense of the live feeds.

Although there are no studies that have tested the palat-

ability of diets formulated using C. nilotica, results by Liti

et al. (2006) provided a strong evidence that the food ingre-

dient as the main protein sources encouraged better assimi-

lation of the diet when included in the diet for Oreochromis

nilotica in earthen ponds. When we compared the ingestion

of Artemia in treatments involving the conventional com-

mercial dry diet of C. gariepinus and in treatments with

formulated diets, we obtained up to 20% reduction in the

Artemia ingestion in treatments involving co-feeding. This

point to the preference of formulated dry diets than the

conventional commercial feed, which enhanced the con-

sumption of more formulated dry diet and a concomitant

reduction in Artemia ingestion. This apparent reduction in

live feed consumption after only 14 days of weaning could be

a reasonable option for many catfish farmers to use suitable

formulated dry diets during weaning.

Current larval growths were higher than those reported by

Awaiss & Kestemont (1998) and Verreth & Van Tongeren

(1989). Improved growth performance of larvae co-fed dry

diets in combination with live feeds was attributed to cou-

pling of nutrients and energy from the Artemia and formu-

lated diet. In gold fish, Abi-Ayad & Kestemont (1994)

established that mixed diets provided growth and survival

results that were intermediary between those obtained from

exclusive live or dry diets during the first week after hatch-

ing. Comparatively, higher growth performance of C. gari-

epinus larvae co-fed formulated diet in combination with live

feeds was attributed to high protein content in the diet

(Table 1) and presence of essential amino acid, gamma

linoleic acid profile, in addition to variable quantities of

vitamins in the diet often associated with C. nilotica (Liti

et al. 2006). The diet was formulated using freshwater atyid

shrimp, C. nilotica, which contains high protein content (66–

72%) and is currently a major food source for wild fish

(Witte et al. 2006). Weaning strategies involving direct

feeding in dry diets resulted in poor growth performance.

One possible reason could be attributed to inability of

C. gariepinus to assimilate higher quantity of dry diets

probably related to slow pace of physiological gut matura-

tion or complete development of the digestive system (Hecht

& Appelbaum 1987; Verreth & Den Bieman 1987; Verreth &

Van Tongeren 1989; Verreth 1994; Awaiss & Kestemont

1998; Kolkvoski 2001). Because of the scanty information

about detailed anatomical and physiological description of

the larval ontogeny of C. gariepinus (Hecht & Appelbaum

1987), the length of the larval period or the right time by

which the species has fully differentiated gut to totally han-

dle dry diets is unclear. Because growth trend curves in the

present study differentiated at different sizes in C. gariepinus

larvae during the growth trial periods, it suggests an exis-

tence of differential critical standing crops (the point at

which growth declines for each individual) for each feed.

Formulated dry diets containing C. nilotica as the main

protein ingredient become efficient in sustaining growth of

C. gariepinus than commercial diet after 1 week of weaning

when the larvae had attained a mean weight of about 45 mg.

Considering that catfish in the wild feed on C. nilotica, it is

possible that they also have particular preference for the

C. nilotica when incorporated into the formulated diet.

Evidence for preference of raw C. nilotica by O. niloticus was

adduced by Witte et al. (2006), yet research on the feed

formulation using this ingredient is scanty.

Although survival has never been a major concern in the

culture of C. gariepinus because it is resistant to water quality

stress as well as common diseases (Hogendoorn 1979; Haylor

1991; Kestemont et al. 2007), use of poor feeding strategies

are major sources of mortalities in larval stages of this

species. Feeding on Artemia as well as combination of

Artemia and dry diets did not compromise survival when

compared to feeding dry diets alone in agreement with

Awaiss & Kestemont (1998) who combined freshwater

rotifers (Branchionus) with dry formulated diets. This is

associated with physiological immaturation of the stomach

to handle dry diet. This is precursor of the importance of

providing lower quantity dry feeds in presence of live feeds to

juvenile C. gariepinus. It would be advisable to avoid feeding

dry diets in absence of live feeds. Such strategy not only leads

to mortality but also encourages wastage of feeds, which

increases the overhead costs of larval nutrition.

Better nutrient utilization parameters occurred when

feeding was carried out using 50% dry diets in combination

with supply of live feed. Comparatively, they were better

than those reported by Awaiss & Kestemont (1998). This

seems to point to the higher intake efficiency of formulated

diets in combination with live feeds. This difference in

nutrient utilization efficiency, therefore, stem from inherent

satisfaction of energy demands of the larvae. It is also logical

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

to suggest that better nutrient utilization was a direct

function of digestibility in larval fish when the combination

of formulated diet and Artemia were provided. Although

current study did not determine digestibility, Letcher (1990)

previously showed that larvae of most fish exhibited high

assimilation when combination of live diets and dry feeds is

provided. High digestibility of the current feed seems to be

related to the nutritional properties of both the dry and live

diets, whose coupling effects encouraged positive nutrient

utilization strategies. Mainly because live diets contain

exogenous substances such as gut neuropeptides, enzymes

and nutritional growth factors that contribute to digestive

efficiencies (Rosenlund et al. 1997; Kolkvoski 2001), while

formulated diet provided more proteins from C. nilotica. The

combinations of the above are likely to improve nutrient

utilization efficiency that enhanced growth even when lower

ration of the diets were supplied.

Early larval feeding has been a limiting factor in the

development of African catfish farming because of the con-

tinuous reliance on mainly Artemia feeds, which is increas-

ingly becoming more expensive and requires specialized

facilities to produce; besides most farmers have no technol-

ogy to quantify the amount of live feeds consumed by the

larvae. The current study trial advanced the use of formu-

lated dry feeds in combination with live feeds, which enabled

up to 20% reduction in live feed consumption after only

2 weeks of weaning. At the same weaning strategy, growth

performance, survival and nutrient utilization were

improved than the use of Artemia alone as well as using

available commercial catfish diet. In this study, inclusion of

C. nilotica ensured high protein in the diet equivalent to the

protein contents in most trout, and salmon starter feeds

seemed to enhance the nutritional quality of the formulated

diet. However, even with such high protein level, the for-

mulated diet resulted in higher mortality when fed to the fish

without live feeds and therefore such weaning strategies must

be avoided. The success of enhancing weaning with more

protein feed ingredients will encourage feed manufacturers

to continue on improvement of larval feed qualities. The

present study therefore provides new insights that will

enhance formulation of feeds using local protein sources that

are currently underutilized. Caridina nilotica being a

by-catch in the fishery and currently not utilized by human

and with high protein content suitable for larval fish should

be used in larval nutrition. Although C. nilotica is still a

by-catch, new research into ways of culturing this freshwater

species will hold future promise for mass production of the

food that will be useful for commercial formulation of fish

feeds.

We would like to acknowledge the financial support given

by Aquaculture Collaborative Research Support Program

(ACRSP) partially funded by the United States Agency for

International Development (USAID) under Grant No.

LAG-G-00-96-90015-00. The authors also thank Stephen

Njau, Beatrice Wambui and Hellen Chebet for providing

the test animals and feeding of the larvae as well as assis-

tance in hatchery management during the entire experi-

mental period.

Abi-Ayad, A. & Kestemont, P. (1994) Comparison of nutritional

status of goldfish larvae Carassius auratus fed with live, mixed or

dry diets. Aquaculture, 128, 163–176.

Adewolu, M.A., Adeniji, C.A. & Adejobi, B. (2008) Feed utilization,

growth and survival of Clarias gariepinus (Burchell 1822) finger-

lings cultured under different photoperiods. Aquaculture, 283,

64–67.

Awaiss, A. & Kestemont, P. (1998) Feeding sequence (Branchionus

and dry diet) on survival, growth and biochemical composition of

African catfish, (Clarias gariepinus, Burchell Pisces: Clariidae)

larvae. Aquac. Res., 29, 731–741.

Gonzalez, A., Celada, J.D., Gonzalez, R., Garcıa, V., Carral.,

J.M. & Saez-Royuela, M. (2008) Artemia nauplii and two for-

mulated replacements as dietary supplement for juvenile signal

crayfish, Pacifastacus leniusculus, Astacidae, from the onset of

exogenous feeding under controlled conditions. Aquaculture, 281,

83–86.

de Graaf, G.J., Galemoni, F. & Banzoussi, B. (1995) The artificial

reproduction and fingerling of the African catfish, Clarias

gariepinus, (Burchell 1822) in protected and unprotected ponds.

Aquac. Res., 26, 233–234.

Haylor, G.S. (1991) Controlled hatchery production of Clarias

gariepinus, Burchell, 1822: growth and survival of larvae at high

stocking density. Aquaculture, 22, 405–422.

Hecht, T. & Appelbaum, S. (1987) Notes on the growth of Israel

sharptooth catfish (C. gariepinus) during the primary nursing

phase in glass aquaria. Aquaculture, 63, 195–204.

Hogendoorn, H. (1979) Controlled propagation of the African cat-

fish, Clarias lazera): reproductive biology and field experiments.

Aquaculture, 17, 323–333.

Kamler, E., Lewkowicz, S., Uchmanski, J. & Urban-Jevierska, E.

(1986) Gravimetric techniques for measuring consumption of

live feeds and artificial diets by fish larvae.Aquaculture, 54, 109–122.

Kerdchuen, N. & Legendre, M. (1994) Larval rearing of an African

catfish, Heterobranchus longifilis (Teleostei, Clariidae): a compar-

ison between natural and artificial diet. Aquat. Living Resour., 7,

247–253.

Kestemont, P., Toko, I., Fiogbe, E.D. & Koukpode, B. (2007)

Rearing African catfish (Clarias gariepinus) and vundu catfish

(Heterobranchus longifilis) in traditional fish ponds (whedos):

effects of stocking density on growth, production and body

composition. Aquaculture, 262, 65–72.

Kolkvoski, S. (2001) Digestive enzymes in fish larvae and juveniles:

implications and application to formulated diets. Aquaculture, 200,

181–201.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd

Letcher, B.H. (1990) Laboratory Growth Rate, Daily Food

Consumption Patterns by Larval Inland Silversides, Menidia beryl-

lina; An Investigation Using Image Analysis. Masters thesis.

University of Rhode Island, Kingston, RI, USA.

Liti, D.M., Waidbacher, H., Straif, M., Mbaluka, R.K., Munguti,

J.M. & Kyenze, M.M. (2006) Effects of partial and complete

replacement of fresh water shrimp meal (Caridina nilotica, Roux)

with a mixture of plant protein sources on growth performance of

Nile tilapia (Oreochromis niloticus L.) in fertilized ponds. Aquac.

Res., 37, 177–483.

Rasowo, J., Ngugi, C.C. & Macharia, M. (2008) Comparative study

of the growth performance of African catfish Clarias gariepinus

(Burchell) fingerlings fed diets based on fresh water atyid shrimp

meal, dagaa fish meal and soybean meal protein sources. E. Afr. J.

Pure Appl. Sci., 1, 1–6.

Rosenlund, G., Stoss, J. & Talbot, C. (1997) Co-feeding marine fish

larvae with inert and live diet. Aquaculture, 155, 183–191.

Tacon, G.J.A. (1990) Standard Methods for the Nutrition and Feeding

of Farmed Fish and Shrimp. Argent Laboratories Press, Redmond,

WA, USA. 454 pp.

Verreth, J. (1994) Nutrition and Related Ontogenetic Aspect in Larvae

of the African Catfish, Clarias gariepinus. Ph.D Thesis, University

of Wageningen. Wageningen.

Verreth, J. & Den Bieman, J. (1987) Quantification of food

requirements of African catfish (Clarias gariepinus Burchell) larvae

fed with decapsulated cysts of Artemia. I. The effects of tempera-

ture and feeding levels. Aquaculture, 63, 251–267.

Verreth, J. & Van Tongeren, M. (1989) Weaning time in Clarias

gariepinus (Burchell) larvae. Aquaculture, 83, 81–88.

Verreth, J., Storch, V. & Segner, H. (1987) A comparative study on

the nutritional quality of decapsulated Artemia cysts, micro-

encapsulated egg diets and enriched dry feeds for Clarias gariepi-

nus, (Burchell 1859) larvae. Aquaculture, 63, 269–282.

Verreth, J., Eding, E.H., Rao, G.R.M., Huskens, F. & Segner, H.

(1993) A review of the feeding practices, growth and nutritional

physiology in larvae of the catfishes Clarias gariepinus and Clarias

batrachus. J. World. Aquac. Soc., 24, 135–144.

Witte, F., Goudswaard, K.P.C. & Wanink, J.H. (2006) The shrimp

Caridina nilotica in Lake Victoria (East Africa), before and after

the Nile perch increase. Hydrobiologia, 563, 31–44.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Aquaculture Nutrition 17; e82–e89 � 2010 Blackwell Publishing Ltd