\  COM PARATIVE STUDY OF THE REPRODUCTIVE AND EARLY LIFE G RO W TH  
PERFORMANCE OF THREE STOCKS OF THE AFRICAN C AT FISH , CLARIAS  
GARIEPINUS, (BURCHELL, 1822) IN GHANA.
By
TEYE CHARLES 
(10256428)
BSc. Agriculture Technology (UDS)
THIS THESIS IS SUBM ITTED TO THE UNIVERSITY OF G H ANA, LEGO N IN 
PARTIAL FULFILLM ENT OF THE REQUIREM ENT FOR THE AW ARD  OF M PHIL
FISHERIES SCIENCE DEGREE
M ARCH, 2011
DECLARATION
I, Teye Charles do hereby certify that, this is the result of my personal work submitted as 
a thesis for a degree in this University and that no previous submission for a degree has 
been done here or elsewhere. References made herein are duly acknowledged.
\v ................
TEYE CHARLES
STUDENT PRINCIPAL SUPERVISOR
DATE: /.3u>AI D A T E :....... ,P.3>. (
MR. D. K. ATSU 
SUPERVISOR 
DATE:
DEDICATION
I dedicate this work to my lovely family.
ACKNOWLEDGEMENTS
My indebtedness goes to my dynamic and industrious supervisors, Prof. P.K. Ofori 
Danson and Mr. D.K. Atsu both of the Department of Fisheries and Oceanography, 
University of Ghana for their pieces of advice, constructive criticisms and inestimable 
contribution to this work in spite of their tight schedules. May God abundantly bless 
them.
1 sincerely register my profound gratitude to Dr. J.N. Padi, the Deputy Officer-In-Charge 
at Aquaculture Research and Development Centre (ARDEC) at Akosombo for his 
technical advice and support throughout the study period. I really appreciate it. Thanks a 
lot.
I am particularly grateful to Mrs. Patience Atsakpo, a Laboratory Technician at ARDEC 
for assisting me in analyzing all the water quality parameters taken during the study 
period.
I cannot but to thank all the Staff of ARDEC for their invaluable support throughout the 
study period.
Special mention must be made of my lovely family for their immense assistance. I love 
you all.
Finally, I wish to thank my colleagues and all and sundry who have in one way or the 
other provided assistance towards the success of this work. Stay blessed.
TABLE OF CONTENTS
TITLE PAGES
D^  eclarat. ion............................................................................................................. ................ ii
Dedication....................................................................................................................................... 111
Acknowledgements........................................................................................................................1V
Table of contents............................................................................................................................. v
List of Tables................................................................................................................................. *x
List of Figures................................................................................................................................. x
List of Plates.................................................................................................................................. xi
ABSTRACT............................................................................................................................................ 1
CHAPTER ONE:
1.0 INTRODUCTIO N................................................................................................................. ..
1.1 Justification.................................................................................................... g
1.2 Objectives..................................................................................... -j
1.3 Research Hypothesis.................................................................. g
iv
CHATER TWO:
2.0 LITERATURE REVIEW .............................................................................................. .9
2.1. Taxonomy and Morphology..............................................................................................9
2.2. Reproduction of C. gariepinus......................................................................................10
2.3. Early life growth performance of C. gariepinus........................................................ 18
CHAPTER THREE:
3.0 M ATERIALS AND M ETHODS................................................................................... 23
3.1 Study area........................................................................................................................23
3.2 Collection of fish Samples............................................................................................23
3.2.1 Broodstock Management..............................................................................................24
3.3 Artificial Reproduction (Hypophysation)....................................................................24
3.3.1 Hormone injection..........................................................................................................28
3.3.2 Maturation Processes, Stripping and Incubation of fertilized eggs........................ 29
3.4 Fry nursing in earthen ponds......................................................................................... 31
3.4.1 Pond preparation and Stocking...................................................................................... 3 j
3.4.2 Daily supplementary feeding...............................................................................  32
3.4.3 Sampling of fry......................................................................... 32
3.5. Reproductive and Growth Parameters................................................... 39
3.5.1 Fecundity.........................................................
V
3.5.2 Gonado-Somatic Index (GSI)
3.5.3 Hatchability..................................................................................... ................33
3.5.4 Survival Rate.......................................................................................... ....... 33
3.5.5 Specific Growth Rate (SGR).......................................................................................
3.5.6 Weight gain.....................................................................................................................
3.5.7 Condition Factor............................................................................................................ ^
3.6 Water Quality Parameters.............................................................................................35
3.6.1 Temperature Dissolved Oxygen and pH.....................................................................36
3.6.2 Turbidity......................................................................................................................... 36
3.6.3 Ammonia (NH3-N).........................................................................................................36
3.6.4 Nitrite (NO2-N).............................................................................................................37
3.6.5 Nitrate (NO3-N).............................................................................................................37
3.7 Data Analysis................................................................................................................. 38
CHAPTER FOUR:
4.0 RESULTS...............................................................................................................................3 9
4 .1 Reproductive Characteristics of C. gariepinus...........................................................39
4.1.1 Relationship between fecundity, gonad weight and fish weight..............................42
4.2. Growth Parameters........................................................................ 45
4.2.1 Condition Factor................................
4.2.2 Length-Weight relationship .46
4.3 Water quality Parameters ... .48
CHAPTER FIVE:
5.0 DISCUSSIO N.........................................................................................................................4 9
CHAPTER SIX:
6.0 CONCLUSIONS AND RECO M M ENDATIO NS................................................... 55
6.1 CONCLUSIONS..................................................................................................................55
6.2 RECOM M ENDATIONS.................................................................................................. 57
REFERENCES.................................................................................................................................... 58
APPENDICES......................................................................................................................................67
Appendix 1: Six Chronological stages seen within the development
of oocyte o f C. gariepinus...............................................................................67
Appendix 2: Analysis of Variance (ANOVA) of Reproductive and Growth
performance of C. gariepinus at 95% confidence level...............................70
LIST OF TABLES r
Table 4.1: The stock, sex, average fish weight, average gonad weight, 
average number o f eggs, hatchability and the
Gonado-Somatic Index of C. gariepinus.............................................................41
Table 4.2: The locality, number of eggs per female of
C. gariepinus and the A uthors................................................................................42
Table 4.3: Mean Relative growth and survival of three
stocks of C. gariepinus...........................................................................................45
Table 4.4: Water quality parameters taken during the culture period
for stocks A, P and K ................................................................................................48
ix
LIST OF FIGURES PAGES
Fig 3.1: The study area, ARDEC, Akosombo ........ 25
Fig.4.1: Relationship between fecundity and fish weights of
the three stocks of C. gariepinus.............................................................................. 43
Fig.4.2: Relationship between fecundity and gonad weight
of the three stocks of C. gariepinus........................................................................44
Fig.4.3 Logarithmic Length-Weight relationship of ARDEC stock, Pacific stock
and Kumah stock.......................................................................................................47
X
LIST OF PLATES
Plate 1: Experimental ponds used for the study at ARDEC, Akosombo.......................... - 6
Plate 2: Feeding of fries stocked in pond................................................................................
Plate 3: Sexing of C. gariepinus.............................................................................................. 11
Plate 4: Some of the Broodstock used.................................................................................... -?
Plate 5: Concrete tank used to hold fish..................................................................................27
Plate 6 : Removal of pituitary gland........................................................................................ 27
Plate 7: Injection of female broodstock..................................................................................27
Plate 8 : Testes of a male broodstock....................................................................................... 27
Plate 9: Stripping of eggs of the injected female.................................................................. 27
Plate 10: Eggs in hatchery bowls..............................................................................................27
Plate 11: Fries kept in bowls..................................................................................................... 27
Plate 12: Some fmgerlings harvested....................................................................................... 27
xi
1
ABSTRACT
A comparative study was carried out into the reproductive and early life growth 
performance of three stocks of the African catfish, Clarias gariepinus (Burchell 1822) in 
Ghana under pond culture system at Aquaculture Research and Development Center 
(ARDEC) in Akosombo in the Eastern Region from January, 2009 to October, 2009. 
Broodstocks of C. gariepinus were collected from Kumah Farms in Kumasi (Ashanti 
Region), the Pacific Farms in Ashaiman (Greater Accra Region) and ARDEC at 
Akosombo (Eastern Region). The broodstocks were used for artificial propagation under 
more controlled conditions including; stripping of eggs, collection of the sperm, followed 
by fertilization and incubation of eggs. The fries were stocked at 2000 fries per 200 m2 
pond (10 fries per lm 2) in six fenced experimental ponds at an average weight of 0.003 g 
and harvested at an average weight of 3.304 g after six weeks pond culture for five 
consecutive times. Differences in fecundity (P = 0.592) and gonado-somatic index (P = 
0.114) as well as specific growth rate (P = 0.163), condition factor (P = 0.345) and 
survival rates (P = 0.601) among the three stocks were all not statistically significant. 
Temperature, Dissolved Oxygen and pH ranges from 30.88±0.75 to 31.15±0.87,
4.28±2.34 to 4.99±2.45 and 6.18±0.15 to 6.98±0.61 respectively. These water quality 
parameters taken were all within the conducive ranges for the culture of the fish species. 
No statistical difference was found among the mean weight of ARDEC stock (3.496 g). 
Pacific farms stock (3.304 g) and Kumah farms stock (3.256 g) indicating that C. 
gariepinus grows at approximately the same rate regardless of its geographical location. 
This implies that fish farmers can reduce cost and save time and resources as they rely on 
the nearest fish seed (fingerlings) for stocking.
CHAPTER ONE
1.0 INTRODUCTION
In recent years the culture of species of the catfish belonging to the family Clariidae is 
fast gaining global attention. C. gariepinus is probably the most widely distributed 
freshwater fish in Africa (Skelton, 1993). The major species of catfish that are of 
commercial interest and cultured in Africa include C. gariepinus, Heterobranchus spp 
and their hybrids. However, comparison of growth rates according to Skelton (1993) 
indicate that C. gariepinus performs better in terms of growth rates and feed conversion 
than other species of the genus Clarias. FAO (2005) emphasized that aquaculture in 
Ghana focusses mainly on Nile tilapia, O. niloticus and C. gariepinus.
Out of the cultured fish species worldwide, C. gariepinus is highly favoured owing to the 
following qualities: high growth rate, hardy and resistant to handling and stress, disease 
resistant, highly valued predominant food fish, much appreciated in a wide number of 
African countries, highly fecund and easily spawned under captive conditions, 
amenability to high density culture, related to their air-breathing habits, it is omnivorous 
and used as a population control check for Tilapias in a polyculture system (Huisman and 
Richter, 1987; Haylor, 1993; De Graaf and Janssen, 1996).
C. gariepinus has almost Pan-African distribution, ranging from the Nile to West Africa 
and from Algeria to Southern Africa and widespread in the tropics. Having evolved in the
3
Pliocene epoch (upper Tertiary period) some 7-10 million years BP, it can tolerate 
salinities up to 2.2 ppt (Clay, 1977).
Aquaculture potential of C. gariepinus in Africa was first realized by Douglas He> at the 
Jonkershoek Fish Hatchery in the Western Cape Province in South Africa in 1941 
(Richter, 1979). It has since been considered to be a fish of great promise for fish farming 
in Africa. This fish species was later confirmed as one of the most suitable species for 
aquaculture in Africa (CTFT, 1972; Micha, 1973; Jocque, 1975; Pham, 1975, 
Hogendoom, 1979; Richter, 1979).
C. gariepinus, is also of a well-grounded position in European fish culture (Huisman and 
Richter, 1987; Kuczyriski et al., 1999), such experiments were carried out by numerous 
authors (among others by Eding et al., 1982; De Leeuw et al., 1985; Richter et al., 
1985,1987; Goos et al., 1987; Kouril et al., 1992; Inyang and Hettiarchichi, 1994). It is 
also valuable for aquaculture worldwide (Khwuanjai Hengsawat et al., 1997).
C. gariepinus spawn naturally in floodplains during the rainy season, and this spawning 
is induced by rise in water levels (Pillay, 1990). However, seed collection from the wild 
is unreliable and limited to the rainy season. In captivity the African Catfish does not 
spawn spontaneously since environmental factors such as the rise in water level and 
inundation of shallow areas do not occur on the fish farms (De Graaf and Janseen, 1996). 
A combination of physical, chemical and biological factors, such as changes in water 
level, pH, water temperature, clarity and flow rate, flooding of marginal plants,
4
associated chem ical changes, and access to suitable spaw ning sites, are responsible tor 
triggering the spawning o f  catfish w hich hardly reproduces in captivity (H ow erton, 
2001).
C. gariepinus has been reared for almost 20 years in Africa with mixed success; the total 
farm production of this species being only 3,978 metric tonnes or 7.4% of the total 
farmed fish production of 69,434 mt in Africa in 1994. To a large extent the poor 
performance of this freshwater fish species in Africa has been due to the absence of 
reliable production techniques for the reproduction and rearing o f the species under 
practical farming conditions (De Graaf and Janseen, 1996).
The development of a reliable method for the production of C. gariepinus fmgerlings was 
one of the priorities of aquaculture research in Africa (Anonymous, 1987). Hormone- 
induced reproduction of the African catfish using deoxycorticosterone acetate, human 
chorionic gonadotropin and common carp pituitaries has been carried out successfully 
(Hogendoom and Vismans, 1980; El Bolock, 1976; Hogendoom and Wieme, 1976; 
Micha, 1976).
Since the early 1970's several techniques have been developed (with or without hormone 
treatment) for the artificial reproduction of the African catfish. Semi natural or hormone 
induced reproduction within ponds or concrete tanks can be used on small farms to 
produce their own larvae and fmgerlings. However, the method has not proved to be a
5
reliable method for mass production needed for larger fish farms or distribution centers of 
catfish fingerlings (De Graaf et al., 1995)
Artificial propagation of C. gariepinus under more controlled conditions including, 
stripping of eggs, collection of the sperm, followed by fertilization o f eggs is now carried 
out in hatcheries with hormonal induction. Artificial reproduction by induced breeding 
through hormone treatment followed by artificial fertilization and incubation of fertilized 
eggs and the subsequent rearing to fingerling size has several advantages as compared to 
the natural reproduction (Woynarowich and Horvath, 1980).
These advantages include better rates of fertilization and hatching, protection against 
predators and unfavourable environmental conditions and better conditions for growth 
and survival. Hogendoom (1980) and Hogendoom and Vismans, (1980) successfully 
developed an intensive production system for African catfish fingerling production based 
on the use of Artemia salina nauplii as feed.
The introduction of intensive rearing methods in the Central African Republic and the 
Ivory Coast encountered numerous technical and economic problems (Janssen, 1985a, 
1985b and 1985c, De Graaf, 1989). The main problem of fingerling production within 
ponds was fish survival rate which was unreliable and varied between 0 - 60 
fingerlings/m2/cycle (Micha, 1973, 1976; Hogendoom and Wieme, 1976; Hogendoom. 
1979) or between 0 and 90 % (Micha, 1975; Janssen, 1985a).
L
6
In the late 1980's, a simple and reliable method developed by De Graaf et al. (1995) in 
Congo-Brazzaville for the nursing of C. gariepinus within protected ponds revealed that 
competition for feed and cannibalism were the major factors affecting pond nursing.
1.1 JUSTIFICATION
Despite the popularity of C. gariepinus as a species for culture and its great market 
potentials, production is still basically at subsistence level due largely to inadequate 
availability of seed for stocking and feed problems. In Ghana, the fmgerlings supplied 
from both the government and privately owned hatcheries are not enough to meet the 
catfish farmers’ fmgerling demands as such its market demands far exceeds supply. 
There is therefore an urgent need to increase the supply of seed through the development 
of simple and efficient seed production and management protocols that are easy to adopt 
by small-scale fish farmers themselves.
In fish reproduction under controlled conditions attempts are made to obtain eggs o f the 
highest weight possible and of the best quality, and hence to produce the highest possible 
numbers of good quality hatch. Up till now, there is lack o f adequate scientific 
information on the reproductive and early life growth performances of different stocks of 
C. gariepinus in Ghana.
This study was therefore geared towards comparing the reproductive and early life 
growth performance of three stocks of C. gariepnus in Ghana as baseline information that 
will be employed to genetically improve the fish species in Ghana.
1.2 OBJECTIVES
The main objectives of this study were to determine and compare the reproductive 
capacity of broodstocks of C. gariepinus from three different populations and the early 
life growth performance of their fries in Ghana.
The specific objectives were to:
• compare the fecundity, hatchability and gonado-somatic index of the broodstocks 
from Kumah farms (Kumasi), Pacific farms (Ashaiman) and ARDEC 
(Akosombo).
• compare the growth rate, survival rate, condition factor and specific growth rate 
of the fries of the broodstocks from Kumah farms (Kumasi), Pacific farms 
(Ashaiman) and ARDEC (Akosombo).
• assess the ambient water quality environment for culture of the fries of the 
broodstocks from Kumah farms (Kumasi), Pacific farms (Ashaiman) and 
ARDEC (Akosombo).
1.3 RESEARCH HYPOTHESIS
• H0: Growth rate of the fries of the broodstocks o f C. gariepinus from Kumah 
farms (Kumasi), Pacific farms (Ashaiman) and ARDEC (Akosombo) in pond 
culture system in Ghana is not significantly different.
• Hi: Growth rate of the fries of the broodstocks of C. gariepinus from Kumah 
farms (Kumasi), Pacific farms (Ashaiman) and ARDEC (Akosombo) in pond 
culture system in Ghana is significantly different.
CHAPTER TWO
2.0 LITERATURE REVIEW
2.1 Taxonomy and Morphology
C. gariepinus belongs to the Kingdom Animalia, Phylum Chordata, Subphylum 
Vertebrata, Order Siluriformes, Family Clariidae, Genus Clarias, Species gariepinus, 
Scientific name: Clarias gariepinus. Based on morphological, anatomical and 
biographical studies carried out by Teugels (1982a, 1982b, 1984), six species were 
identified in the freshwaters of Ghana as follows: C. gariepinus (Burchell, 1822), C. 
anguillaris (Linnaaeus, 1758), C. laeviceps (Gill, 1863), C. camerunensis (Lonnberg, 
1922), C. ebriensis (Pellegrin, 1922), C. agboyiensis (Sydenham, 1980).
The Clarias genus can be described as displaying an eel shape, having an elongated 
cylindrical body with dorsal and anal fins being extremely long (nearly reaching or 
reaching the caudal fin) both fins containing only soft fin rays. The outer pectoral ray is 
in the form of a spine and the pelvic fin normally has six soft trays. The head is flattened, 
highly ossified, the skull bones (above and on the sides) forming a casque and the body is 
covered with a smooth scaleless skin (De Graaf et al., 1995).
The skin is generally darkly pigmented on the dorsal and lateral parts of the body. The 
colour is uniform marbled and changes from greyish olive to blackish according to the 
substrate. On exposure to light, the skin colour generally becomes lighter. They have four
pairs of unbranched barbels, one nasal, one maxillar (longest and most mobile) on the 
vomer and two mandibulars (inner and outer) on the jaw (De Graaf et al., 1995).
Tooth plates are present on the jaws as well as on the vomer. The major function of the 
barbels is prey detection. A supra-branchial or accessory respiratory organ, composed of 
a paired pear-shaped air-chamber containing two arborescent structures is generally 
present. These arborescent or cauliflower-like structures located on the secondhand forth 
branchial arcs, are supported by cartilage and covered by highly vascularised tissue 
which can absorb oxygen from atmospheric air (Moussa, 1956).
The air chamber communicates with the pharynx and with the gill chamber. The 
accessory air breathing organ allows the fish to survive for many hours out o f the water 
or for many weeks in muddy marshes. Clarias spp. inhabits calm waters from lakes, 
streams, rivers, swamps to floodplains, some of which are subject to seasonal drying. The 
most common habitats frequented are floodplain swamps and pools in which the catfish 
can survive during the dry seasons due to the presence of the accessory air breathing 
organs (Bruton, 1979; Clay, 1979).
2.2 Reproduction of C. gariepinus
Information on the reproduction of C. gariepinus has been given by several researchers in 
Africa (Mulder, 1971; Willoughby and Tweddle, 1978; Bruton, 1979; Clay, 1979; Micha. 
1976; Richter, 1979; Clay, 1981; Nawar and Yoakim, 1984; Quick and Bruton, 1984).
11
Many Clarias species undertake lateral migrations from the larger water bodies in which 
they feed and mature, according to Legendre and Jalabert (1988) at about the age of two 
years under natural conditions (but sometimes 7-8 months in captivity), to temporarily 
flooded marginal areas to breed (Seegers, 1996).
These reproductive migrations typically take place shortly after the onset of the rainy 
season(s). Predictive cues for gonadal maturation include increasing temperature, 
photoperiod and electrical conductivity, while proximate cues for final maturation are 
associated with rising water levels (De Graaf and Janssen 1996; Legendre and Jalabert 
1988).
The adult female C. gariepinus has a fully developed ovary which contains "ripe" eggs 
the whole year through and is capable of reproduction if  kept in ponds at water 
temperature above 22 °C. The oocyte development decreases once the temperature drops 
below 22 °C (Young and Fast, 1990).
Under ideal conditions, the ovary of a mature female constitutes 15-20% of the fish’s
body weight, with about 600 ova per gram of ovulated eggs (De Graaf and Janssen,
1996). Willoughby and Tweddle (1978) stated that the mature ovaries contained between
600 and 1,400 eggs per gram wet weight for C. gariepinus living in the Shire Valley, 
Malawi.
12
In the dry season (June-July-August) the water temperature drops b elow  22 °C and the 
ovary makes up approximately 5% o f  the body w eight o f  the fem ale. Artificial 
reproduction is still possible but the number o f  eggs obtained is sm all and the quality of 
the eggs decreases (D e Graaf and Janssen, 1996).
In general, the testis of a male is fully developed at an age of 8-12 months once the males 
reach a weight of approximately 200 g. In Congo-Brazzaville sperm could be obtained 
the whole year through and no impacts of the temperature on the availability of sperm 
was found (De Graaf and Janssen, 1996). Wootton (1992) stated that the fish reach 
sexual maturity at an unusually small size (in length and weight) in stunted populations. 
Some factors, such as pollution, uncontrolled hunting and other physical and chemical 
factors, may stunt the C. gariepinus population.
Ovarian anatomy in Clarias is typical for teleostians. The testes, on the other hand are 
peculiar, being divided into two parts: an anterior functional testes and a posterior cluster 
of seminal vesicles, the role of which is not precisely understood but which may 
contribute to sperm longevity and motility (Legendre and Jalabert, 1988). In terms of 
gonadal activity, males appear to be less sensitive to temperature fluctuations than 
females (De Graaf and Janssen, 1996).
Prior to mating, males have been reported to compete aggressively for females with 
which they mate in single pairs, with the female swishing her tail vigorously to mix the 
eggs and sperm and distribute the fertilized eggs. The adhesive eggs stick to submerged
13
vegetation, hatch in 20-60 hours, depending upon temperature (De Graaf and Janssen. 
1996).
The yolk sac is absorbed within 3-4 days (De Graaf and Janssen, 1996) and the stomach 
is fully functional within 5-6 days after onset of exogenous feeding (Hecht, 1996). Sexual 
differentiation begins between 10-15 days after hatching (Legendre and Jalabert. 1988). 
The female is the heterogametic sex (Volckaert and Agnese, 1996).
Larvae feed on phytoplankton and grow rapidly in the warm, nutrient-rich floodplain, 
reaching a size of 3-7 grams within 30 days (De Graaf et al., 1995). As flooded marginal 
areas dry up with the end of the rains, juveniles and adults make their way back to deeper 
water. In areas with two rainy seasons, there are usually two reproductive peaks during 
the year, corresponding in intensity to the magnitude of the rains (Young and Fast, 1990).
Egg and larval fish can be mainly divided into 5 stages: (1) Egg phase (Incubation 
period); (2) Prolarval stage (yolk sac stage); (3) Larval stage (absorbed yolk sac but 
incomplete organs); (4) Postlarval stage (absorbed yolk sac and complete organs) and 
after that they will develop to (5) juvenile stage (fry and fmgerling). In this stage they are 
similar to adult fish but smaller and reproductive organs are not matured (De Graaf and 
Janssen, 1996).
The use of high quality gametes from captive fish broodstock is of great importance for 
ensuring the production of viable larvae (Kjorsvik et al., 1990). The fish farming industry
14
has been more focused on the quality of eggs or larvae rather than that ot sperm, even 
though the quality of both gametes may affect fertilization success and larval survival 
(Bozkurt et al., 2006). It is well known that the size of eggs of fish shows considerable 
intra- and inter- specific variation. Even parental fish of the same strain, weight and 
length have eggs that are in different sizes (Bagenal, 1971).
Egg size is primarily determined by the genotype of parents of fish and it is also known 
to be affected by other factors including age and size of the female parent. Gall (1974) 
has shown in studies of hatchery-reared trout that older and heavier females produce 
larger eggs than younger and smaller fish. The availability o f food also affects egg size 
(Springate et al., 1985). Alterations in egg size also occur in batch-spawning fish as the 
season progresses (Bagenal, 1971) and in synchronous spawners as a result of 
photoperiodic modifications of maturation (Bromage et al., 1984).
Sperm quality data are required for successful fertilization. Viable sperm is an essential 
component in any successful animal production operation and the success of reproductive 
process is dependent on a supply of high quality gametes (Cruz-Casallas et al., 2005). In 
addition, sperm quality is a measure of the ability of sperm to successfully fertilize an 
egg. Any quantifiable physical parameter that directly correlates with the fertilization 
capacity ot sperm could be potentially used as a measure of sperm quality. Optimal 
sperm quality is important for effective broodstock management and should be a criterion 
in the selection of male broodstock (Bozkurt et al., 2006).
15
In commercial hatcheries, sperm is often inadequate both in terms o f quantity and quality 
and it does not always give successful fertilization in the artificial insemination 
procedures commonly used for aquaculture species. Spermatozoa motility is the most 
reliable indicator used to evaluate semen quality. However, investigations of the 
relationship between spermatozoa motility and fertility have given conflicting results or 
were of little value in predicting fertilization potential (Bozkurt, 2006).
Subjective estimation of motility requires considerable experience and this indicator is 
being used in the selection of sperm for insemination and preservation. Spermatozoa 
motility varies in vigor and duration not only among males but also within an individual 
male depending on its ripeness (Graham et al., 1978). The highest motility o f the 
spermatozoa is observed at the peak of the breeding season (Temer, 1986). Studies on 
most fish species recorded that motility of spermatozoa may show seasonal variation 
(Benau and Temer, 1980).
Fecundity which is the number of ripening eggs found in the female prior to the spawning 
act (Bagenal, 1978) varies greatly in individuals of one species o f the same weight, length 
and age, but in many species it increases in proportion to the size o f fish (Lowe- 
McConnell, 1975). Fecundity is known to be affected by stress and nutritional deficiency 
(Hogendoom and Vismans, 1980).
Most studies of reproduction tend to consider fecundity and egg size as separate 
indicators of reproductive performance. It is generally accepted that there is an inverse
relationship between fecundity and egg size in which fish produce either more eggs oi a 
smaller size or fewer eggs of a larger size (Springate et al., 1985; Bromage et al., 1992).
Egg size is more important than fecundity for the fertilization success. During incubation, 
the eggs should be well oxygenated for maximum hatching to occur (Hogedoom, 1980; 
Viveen et al., 1985; De Graaf and Jansen, 1996). Moreover, the poor survival of eggs and 
larvae is mainly the result of inadequate nutrition during the nursing phase and careless 
nursery management practices. Successful hatching of fish eggs and careful feeding of 
larvae during the early stages of their development is essential for better survival of 
larvae.
There is strong evidence that the degree of stress; the feeling o f pain and fear (Broom, 
1998) in female affect the quality of eggs in induced spawning (Pickering et al., 1995). In 
fish, stress response to different stimuli is well documented (Barton and Iwama, 1991; 
Iwama et al., 1997; Wendelaar-Bonga, 1997; Iwama et al., 1999; Ruane, 2002). 
Wendelaar-Bonga (1997) stated that stress is a condition in which the dynamic 
equilibrium of animal organisms, called homeostasis, is threatened or disturbed as a result 
of the actions of intrinsic or extrinsic stimuli, commonly defined as stressors.
There is no parental care for ensuring the survival of the catfish offspring except by the 
careful choice of a suitable site. Development of eggs and larvae is rapid and the larvae 
are capable of swimming within 48-72 hours after fertilization at 23-28 °C. The 
development of the oocytes of the African catfish is mostly related to temperature (as is
common with a large number offish  species). According to Owiti and Dadzie, (1989), 
six chronological stages can be seen within the development of oocyte (Appendix 1).
For hormone induced reproduction (semi artificial or artificial) the following hormones 
are generally used (De Graaf et al., 1995):
• DOCA (Desoxycorticosteroid Acetate), 2.5-5 mg per 100 gram of female. A 
disadvantage of using this hormone is that it is mostly suspended in oil which 
causes severe ulcers on the injected female.
• HCG (Human Chorionic Gonadotropin), 25 I.U. per 100 gram of female. This 
hormone works well but it is expensive.
• Common carp (Cyprinus carpio) pituitary gland extract, 3-4 mg per kilogram of 
female or 1-2 whole pituitaries per female. In general the common carp pituitary 
gland material has to be imported from abroad which means that it is usually not 
accessible for small fish farms.
• Pituitaries of C. gariepinus. A female catfish will respond once it injected with a 
pituitary of a catfish (male or female) of equal size.
• Pituitaries of the Nile Tilapia (Oreochromis niloticus), 3-4 pituitaries of a Nile 
Tilapia (100 150 gram) per female catfish will induce ovulation.
• Pituitaries of Nile perch (Lates niloticus). 1-2 pituitaries per female catfish will 
induce ovulation.
18
2.3. Early growth performance of C. gariepinus
Growth of fish larvae, postlarvae, fry and fingerling reared in pond, hapa (mosquito 
netting) and tank varies according to species and within species depending on factors 
such as stocking rate, water temperature, feed and feeding, intensity ot light. Growth in 
Clarias species is relatively rapid, with most species approaching their maximum size 
within a couple of years.
Most Clarias species are relatively r-selected (altricial) with medium to high resilience 
(population doubling time of 15-30 months) and rapid growth to relatively early sexual 
maturity, high fecundity and a (anticipated) short lifespan (Hecht, 1996). The author 
further stated that first year growth in the larger, k-selected species (e.g., C. gariepinus) is 
nearly linear resulting in the large initial jump in size (Merona et al., 1988).
Length and weight relationships are of great importance in fisheries research because 
they provide information on population parameters (Garcia et al., 1989; Krause et al., 
1998; Haimovici and Velasco; 2000; Ovredal and Totland; 2002). Length and weight 
measurements in conjunction with age data can give information on the stock 
composition, age at maturity, life span, mortality, growth and production (Beyer, 1987; 
Bolger and Connoly, 1989; King, 1996a and b; Diaz et.al., 2000).
First, a change in length and weight tells the age and year classes of fishes, which is 
important in fishery and Pisciculture. Secondly, the data can be used to estimate the 
mortality rate, and thirdly they can be used to assess the sustaining power of the fishery
19
stock. In addition, the data on length and weight can also provide important clues to 
climatic and environmental changes, and the change in human subsistence practices (Luff 
and Bailey, 2000).
However, the size attained by the individual fish may also vary because of variations in 
food supply, and these in turn may reflect variations in climatic parameters and in the 
supply of nutrients or in the degree of competition for food. Thus, a change in average 
size through a certain period of time may indicate a change in average age resulting from 
those factors (Pauly, 1985). Environmental deterioration, for example, may reduce 
growth rates and may cause a decrease in the average age o f the fish. In reality, the 
interactions between environmental changes and growth rates for instance are believed to 
be complex and difficult to explain (Hecht and Uys, 1997).
Growth is density dependent; the higher the rearing density o f larvae, the lower their 
growth rate. Stocking density of C. gariepinus is considered the most important factor 
affecting cannibalism and aggression. It is known to have a strong influence on growth, 
survival and behaviour of fish (Hecht and Appelbaum, 1988). Increasing the number of 
fish per culture unit is one of the most common practices under farm conditions to 
increase production. Multiple sorting is essential owing to the cannibalistic nature of C. 
gariepinus (Kaiser et al, 1995).
Feeds and feeding of the larvae, fry and fmgerlings of the catfishes have been most 
studied and shown to influence the growth and survival of the fish. Studies have revealed
20
that live zooplankton is the preferred larval food. Many smallholdings merely rear larvae 
to fingerling size in organically fertilized ponds at a density of between 30-1000 
larvae/m2. Fingerlings are stocked into rearing ponds at a rate o f 50-75 fish/m3 under 
good management (Olaleye, 2005).
Davy and Chouinard (1980) noted that the most critical area o f fish fry production and 
the major critical period is immediately before and during the initiation of first feeding. If 
food is not immediately available to fish hatchlings the fry may become weak and 
become predisposed to predation in natural rearing systems (Rana, 1990). Availability o f 
zooplankton during the first week after stocking is essential for the successful nursing of 
the catfish larvae.
Tf  the initial feeding of C. gariepinus fry is delayed beyond 4 - 5days, more than 50% of 
the fish may die (Owodeinde et al., 2004). Huisman et al., (1976) considered the lack of 
suitable food as the main cause of mortality in most fishes at this stage, emphasizing the 
importance of, not only, of the quantity and quality, but also the feed size.
Live food such as Artemia, Daphnia, rotifers, and copepods are the most satisfactory 
“first food” for fry (Bard et al., 1976). After this transition period of two weeks the fry 
can detect and eat artificial food (Madu et al., 1993). Consumption of live food during the 
first four days of feeding ensures adequate survival of C. gariepinus fry (Adeyemo et al.. 
1992). Micha (1973) analyzed the stomach contents of 15-day-old C. gariepinus fry and
21
reported that the entire food contents was zooplankton, and that beyond that age the 
contents changed to larvae of aquatic insects and eventually artificial feed.
In general after four to five weeks after stocking, two main size groups of catfish can be 
recognized within the pond (De Graaf et al., 1995):
• A large group (80-90% of the biomass) consisting of small sized fmgerlings (2-3
g)-
• A small group of fmgerlings (10-20% of the total biomass) consisting of large 
size fmgerlings (8-10 g). Cannibalism will occur (i.e. the large sized fish will eat 
the small ones) if the two groups are not separated and only a very small number 
of large fish will be harvested.
According to De Graaf et al., (1995), three factors are probably influencing fingerling 
production in ponds;
• The presence of amphibian predators like adult frogs which can cause mortality of 
10% due to predation.
• A competition for food resources. Phytoplankton levels, in the unprotected ponds, 
are reduced by the presence of phytophagous frog larvae, consequently leading to 
a reduction in availability of zooplankton, needed by the catfish larvae. This 
effect is critical during the first days of exogenous feeding of the larvae and 
determines the limits of the nursery system which is dependent on natural food 
production. Production levels up to 800 fingerlings/m2 are reported by Hecht et 
al. (1988), the nursing ponds however are stocked with 10 day old larvae which
22
passed the critical moment before which live food or a real “baby” diet (Verreth 
et al., 1991) is essential for survival and production.
.  Sibling cannibalism has been observed earlier among fry o f C. gariepinus in 
hatcheries (Janssen, 1985) and has been studied in detail by Hecht et al. (1988). 
Within aquarium experiments two types of cannibalism can be distinguished; type 
1 or tail first cannibalism, where predator and prey size are almost equal, 
occurring in a weight range o f 0.006 g -  0.9 g and Type 2 or head first 
cannibalism, where predator size exceeds prey size, occurring in a weight range of 
0.9 g -  4.6 g. Once the weight of the fish exceeds 4.6 g, cannibalism ceased to 
exist in these aquarium experiments due to the fact that the mouth width o f the 
largest fish in the sibling population is smaller than the head width of the smallest 
fish. In ponds, two main groups of fmgerlings can be distinguished 40 days after 
stocking: a small group (0-3%) with an weight ranging between 8-12 g, and a 
large group (± 97%) of fmgerlings weighing 0.5 -3  g. It is most likely that type 2 
cannibalism continues to be of importance due to its higher size variation as 
compared with the experiments of Hecht et al., (1988). This phenomenon explains 
why after a long rearing period the number of fry harvested is low and their 
average weight is high.
23
CHAPTER THREE  
3.0 M ATERIALS AND METHODS
3.1 Study area:
The study was conducted in Akosombo (Fig. 3.1) at the Aquaculture Research and 
Development Centre (ARDEC) of the Water Research Institute (WRI) of Council for 
Scientific and Industrial Research (CSIR) in the Eastern Region o f Ghana from January, 
2009 to October, 2009. Akosombo lies at latitudes 06°13'N and longitude 00°04'E. 
ARDEC has aquaculture facilities such as twenty (20) 0.2 ha ponds, twenty (20) 200 m2 
experimental ponds (out of which six (6 ) were used to conduct the study: plates 1 and 2 ), 
twenty (20) 50 m2 ponds, a pumping machine that pumps water from the Volta Lake to 
the ponds, a Laboratory where the analyses of the water quality parameters were done, 
feed formulation machine among others.
3.2 Collection of fish Samples.
Broodstock (gravid females and matured males) of the African Catfish, Clarias 
gufiepiuus, were collected from three populations namely fvuinall Farms Complex in 
Kumasi (Ashanti Region), the Pacific Farms in Ashaiman (Greater Accra Region) and the 
Aquaculture Research and Development Centre (ARDEC) at Akosombo (Eastern 
Region). The Kumah Farms Complex population had its source (the fmgerlings) from
River Tano in Techiman whereas the populations from Pacific Farms and ARDEC had 
their sources from a stream in Ashaiman and the Volta Lake in Dzemeni respectively.
3.2.1 Broodstock M anagement
The broodstocks collected from the fish farms were kept for two months in three separate 
earthen experimental ponds of sizes 200 m2 each and fed twice a day at 9.00 GMT and
15.00 GMT with formulated pelleted feed of 40% crude protein constituting 40% fish 
meal, 35% wheat brown and 25% maize. After the two months’ culture period, the 
matured females of average size 950 g were selected according to the following criteria;
• A well distended, swollen abdomen from which ripe eggs can be obtained by 
slightly pressing the abdomen toward the genital papilla. Ripe eggs are generally 
uniform in size
• A swollen, sometimes reddish or rose coloured genital papilla.
Matured male broodstocks of average size 887 g were also identified using their distinct 
sexual papilla located just behind the anus (Plate 3).
3.3 Artificial Reproduction (Hypophysation)
Broodstocks (Plate 4) from the three Stocks were taken from the ponds every two months 
for five consecutive times. For each time, the broodstocks were held in 1 m2 happas (one 
fish in one happa) fixed in round concrete tanks (Plate 5) to condition them for three
25
THE LOOTtOfc Of THE STUDY «E *
Legend
Buildings 
★ T n e  Study Area 
Roads 
Rivers 
Lakes 
1 Settle m e r»1
Fig. 3.1The study area, ARDEC, Akosombo.
26
Plate 1: Experimental ponds used for the study at ARDEC, Akosombo.
Plate 2: Feeding of fries stocked in pond.
27
Plate 3: Sexing of C. gariepinus Plate 4: Some of the Broodstock used
Plate 7: Injection of Female broodstock Plate 8 : Testes of a male broodstock
Plate 9: Stripping of eggs of the injected female Plate 10: Eggs in Hatchery bowls
Plate 11: Fry kept in bowls Plate 12: Some fingerlings harvested
28
3.3.1 Hormone injection
After the conditioning period, males of approximately the same weight as the females 
from the three Stocks were weighed with an Electronic Scale of Model KERN 572 and 
sacrificed using a sharp knife to cut off the head whilst covering the head with a dry 
towel. The pituitary glands extracted from the dissected heads (Plate 6 ) were ground in a 
portable mortar with a pistil and mixed with a physiological saline solution (9.0 g of 
common salt/litre of water). The testes (milt) of the sacrificed males were preserved in 
the physiological saline solution.
A syringe was filled with 5ml of the suspension of the ground pituitary gland and injected 
intra-muscularly at about 21.00 GMT into the dorsal muscle just below the tail peduncle 
of the female broodstocks (Plate 7) following the method described in De Graaf and 
Janssen (1996) and Viveen et al., (1985). The head of the breeders were covered with a 
towel in order to keep them quiet during the injection of the pituitary hormone. All the 
injected females were returned to the separate hapas fixed in different concrete tanks 
filled with water to enable the breeders to be caught easily the morning after injection so 
as to avoid spoilage of eggs.
The pituitary gland extract triggered the maturation of the eggs of the gravid females. The 
time ot injection until the time of ovulation is temperature dependent (De Graaf and 
Janssen, 1996). The maturation processes of the eggs were completed within 6  to 12
29
hours (a clock was used to monitor the time) at a temperature o f within 26 to 28°C (using 
a thermometer).
3.3.2 Maturation processes, stripping and incubation of fertilized eggs.
The process of final maturation (migration of the nucleus to the animal pole, fusion of the 
yolk, breakdown of the germinal vesicle followed by first meiotic division) and ovulation 
(rupture of the follicles and accumulation of the ripe eggs in the ovary cavity) cannot be 
stopped or reversed after administration of the correct hormone dosage. Once these 
processes start the eggs must either be spawned or stripped. The speed o f the process is 
dependent upon water temperature, the higher the temperature the quicker the eggs 
ovulate.
The males of the African catfish could not be stripped and consequently the sperm was 
only obtained by sacrificing a male. The male was killed and the body surface thoroughly 
dried after which the testis was dissected and placed in a mortar. The testes (Plate 8) of 
the sacrificed males were rapidly cut into small pieces using a pair of scissors. The milt 
was then pressed out of the testes with both hands (Janssen, 1985, De Graaf, 1989) into 
dry plates. The eggs (Plate 9) of the injected females were stripped out into separate dry 
plastic containers by gently pressing their abdomen with a thumb from the pectoral fin 
towards the genital papilla. This was done to avoid direct contact between the eggs and 
water since the latter renders the eggs infertile.
30
Eggs stripped after 6  hours o f injection flowed out easily in a thick je t from the genital 
vent. The ovulated eggs were more or less transparent and flattened. The milt o f  the 
males was sprinkled on the stripped eggs and stirred in order to prevent the eggs from 
sticking together into one clump. The eggs mixed with clean water stored in reservoir 
tanks were stirred continuously for about 60 seconds and then left for about 120  seconds 
before spreading it at the bottom o f the hatchery bowls (Plate 10) with a constant flow o f 
water in order to provide enough oxygen.
After hatching, the larvae (hatchlings) swam to the surface o f  the hatchery bowl where 
they were siphoned from the remaining egg-shells and dead eggs into bigger bowls (Plate 
11) with clean water from the reservoir tanks in order to avoid fungal infections o f 
hatchlings and consequent larval mortalities. This water was changed every 12 hours to 
enable the fry have enough oxygen at all times. The eggs that could not hatch turned 
whitish which is an indication o f its unviability.
The hatchlings were not exposed to direct sunlight since healthy larvae tend to stay in 
dark places in the wild (De G raaf et al., 1995).They were not fed for the first three days 
as they relied on the food resource within their yolk sac. After the three days, the yolk 
sacs were absorbed and the hatchling visibly developed into a fry.
In order to enhance further development and survival o f  the fry at this stage, they were 
fed with live food namely Artemia nauplii for three days after which they were fed 
alternatively with both Artemia nauplii and powdered feed for two days and finally with
31
only powdered feed for additional two days before stocking them in the ponds. This was 
done to introduce them to the powdered feed that they would be fed with during the pond 
culture period.
3.4 Fry nursing in earthen ponds
3.4.1. Pond preparation and Stocking
Six fenced 200 m2 experimental ponds (three stocks with two replicates each) stocked at 
2000 fry per pond with an average fry weight o f  0.003 g were used for the nursing period 
o f fry for five consecutive times at a duration o f  six weeks (42 days) each. Before water 
was pumped from the Volta Lake near ARDEC into the nursery ponds using the Hidel 
pump, grasses were cut and excess silt from the pond bottom removed. The pond bottoms 
were then allowed to dry for few days so as to kill potential fry predators (i.e. water 
insects, amphibian larvae and catfish fmgerlings from previous rearing), and to increase 
the mineralisation (oxidation) o f  nutrients in the pond bottom.
Quicklime (CaO) o f about 3 kg weight was used to lime each nursing pond before 
stocking for the culture period. This was done to disinfect the pond bottom, increase the 
pH o f water and pond bottom to an optimum level (pH 7-9) for plankton and fish 
production and also increase the alkalinity o f  water.
32
3.4.2 Daily supplementary feeding
Daily supplementary feeding o f the fry started a day after stocking with formulated 
powdered feed o f  40% crude protein consisting 60% fish meal, 35% wheat brown and 
5% maize. The feed was ground and sieved with a mosquito net to m uch smaller size that 
could be consumed by the fry. Feed was broadcast on the surface o f  each pond three (3) 
times daily within every 4 hours (8:00 GMT, 12 GM T and 4:00 GM T) at a rate o f  200 g 
per day.
3.4.3 Sampling o f fry
Sampling o f the fry was done after every six weeks culture period. Data on Standard 
Length (SL) was taken to the nearest ± 0 .1  mm using M easuring Board and the W eight 
(W) to the nearest ± 1.0 g using an Electronic Scale of Model KERN 572.
3.5 Reproductive and Growth Parameters
3.5.1 Fecundity
The stripped eggs for each female were weighed with an electronic scale o f  model KERN 
572. Samples (0.5 g) o f eggs was taken in three replicates for each female and counted. 
The total number o f  eggs was estimated from the averages taken as follows:
Fecundity = (Weight o f eggs counted / Total weight o f  eggs stripped) x Average number 
o f eggs counted (Ricker, 1975).
33
3.5.2 Gonado-Somatic Index (GSI)
The Gonado-Somatic Index o f both males and females were calculated by the formula: 
GSI = (Gonad weight / fish body weight) x 100% (Ricker, 1975).
3.5.3 Hatchability
After the hatching period, samples o f both eggs that did not hatch and the hatched fry 
were taken in three replicates from the hatchery bowls. The fry was separated from the 
eggs that did not hatch and both counted. The hatching rates were then calculated by the 
formula:
Hatching Rate (%) = (Average number o f fry counted / Total number o f  both fry and 
eggs) x 100% (Ricker, 1975)
3.5.4 Survival Rate
Survival rate o f  the fry (now fmgerlings) was determined after every final harvesting o f 
the fmgerlings (Plate 12). The total number harvested was counted. The survival rate was 
then computed as:
Survival Rate (%) = (Number o f survivals at the end o f the experiment / number stocked) 
x 100% (Ricker, 1975).
34
3.5.5 Specific Growth Rate (SGR)
Specific Growth rate which is the increase in cell mass ot the fry per unit time is 
expressed as the per cent daily fish body weight gain throughout the culture period. Both 
the final and initial weights o f the fry were taken using an electronic scale o f  model 
KERN 572. The average specific growth rate for each Stock was then calculated as 
follows:
Specific Growth rate (SGR) in %/day = 100 (InW 2- InW i) / change in time (t) (Ricker, 
1975).
Where:
• InW i and InW 2 are natural logarithm o f the initial and final weights (in grams) o f  
the fish respectively.
• Change in time is the culture period (in days) for which fish has grown from Wi 
to W2.
3.5.6 Weight gain
After every culture period o f fry, 200 fish were sampled from each Stock. The average 
weight in grammes was used to calculate the Mean Weight Gain as well as the Mean 
Daily Weight Gains as follows:
Mean Weight Gain = final weight (g) -  initial weight (g) (Ricker, 1975).
Mean Daily Weight Gain (MDWG) = (W2 - W ,)/ (W2 - W ,) x 0.5t (Ricker, 1975).
Where:
.  W! and W2 are initial and final weights (in grams) o f  the fish respectively.
35
• t = culture period (in days) for which fish has grown from W, to W2.
• 0.5 = constant
3.5.7 Condition Factor
The condition factor (K) is a quantitative parameter o f  the w ell-being state o f  the fish and 
reflects recent feeding conditions. The total lengths in cm and weights in grams o f 200 
sampled fish from each Stock after every culture period were taken and used to calculate 
condition factor as follows:
Condition Factor: K = (W / Lb) x 100% (Ricker, 1975).
Where:
• W = weight in grams
• L = total length in (cm)
• b = regression co-efficient (slope) o f  weight / length plots
3.6 Water Quality Parameters
The water quality parameters (physicochemical parameters) o f  the culture medium were 
taken every two weeks.
36
3.6.1 Temperature, Dissolved Oxygen and pH
The temperature (°C), dissolved oxygen (mg/1) and pH o f the pond water for the three 
stocks were measured in situ using the WTM Inolab Oxi Level 2 Oxygen metre for both 
temperature and dissolved oxygen and then Suntex model SP-701 pH meter tor pH. For 
all the three parameters, the probes o f  the measuring instruments were immersed into the 
ponds at a depth o f  about 10 cm at the middle o f  the ponds and the readings taken from 
the meters when equilibrium was attained.
3.6.2 Turbidity.
About 5 ml o f water samples from each pond was poured into a cuvette and inserted into 
the cell holder and the value read from the already calibrated Turbidity metre in 
Nephelometric Turbidity Units (NTU).
3.6.3 Ammonia (NH3-N)
The Direct Nesslerization Method was used to determine Ammonia. A 50 ml o f  each 
water sample was measured into a conical flask. 1 drop (0.05 ml) ethylenediaminetetra 
acetic acid (EDTA) reagent was added and mixed well; 2 ml Nessler Reagent was then 
added and well mixed. A 50 ml zero blank and a 50 ml standard o f 1.00 mg/1 N H 3-N 
were also treated the same way. A reaction process o f at least 10 minutes was allowed 
with the Nessler Reagent. Ammonia-nitrogen's presence was indicated by the yellow 
colouration. The Comspec M201 visible spectrophotometer at a wavelength o f 420 nm
w ith  a 1 cm light path cell or cuvette was used to standardize the zero blank and standard. 
A\mmonia concentration in samples was measured in mg/1 to 2 decimal places.
. .6.4 Nitrite (NO2-N).
Diazotization Method was used in the determination o f  nitrite in samples. A 10 ml o f  
each sample o f  water was measured into a test tube and 1 ml o f  0.3 M sodium hydroxide 
solution added. A 1 ml colouring reagent was also added to each sample. A zero blank 
and standard o f  0.1 mg/1 NO 2-N was used to standardize the visible spectrophotometer at 
a wavelength o f 543 nm. The concentration o f N O 2-N in mg/1 was read w ith a 1 cm light 
path cell after 10 minutes o f colour development. The presence o f a pink colouration 
indicated nitrite-nitrogen.
3.6.5 Nitrate (NO3-N).
Nitrates were determined using the Hydrazine Reduction Method. A 10 ml o f  each 
sample o f water was measured into a test tube and 1 ml 0.3M NaOH (Sodium Hydroxide) 
solution added. A 1 ml reducing mixture was also added. The mixture was then heated 
for 10 minutes at 60 °C and cooled. 1 ml o f colouring reagent was added and 10 minutes 
reaction period was allowed. A zero blank and standard o f 1.00 mg/1 NO 3-N mixture 
were used to standardize the visible spectrophotometer at a wavelength o f  5 4 3  nm with a 
1 cm light path cell. Nitrate concentration in samples was measured in as N O 3-N to 2 
decimal places
38
3.7 Data Analysis
The data on both the growth characteristics and the water quality parameters were 
statistically analyzed using GraphPad 1NSTAT software program (GraphPad Software, 
1993) and presented graphically using M icrosoft excel programme. A One-W ay 
Statistical Analysis o f  Variance (ANOVA) was done for the Female Broodstock weight, 
Male Broodstock weight, Male gonad weight, Egg weight, Fecundity, Gonadosomatic 
Index, Fingerlings weight gained, Specific growth rate, Condition factor and Survival rate 
of the fish from the three stocks using MINITAB. This was to determine whether the 
observed differences were significant.
Data on the fecundity and fish weight, fecundity and gonad weight and length and weight 
o f fish from the three stocks were subjected to Regression and Correlation Analysis at a 
(P<0.05) significance level in other to determine the relationship between them. The 
standard deviation in each growth parameter was calculated and expressed as means ± 
SD.
39
CHAPTER FOUR 
4.0 RESULTS
4.1: Reproductive Characteristics o f C. gariepinus.
The average weights o f brood fishes, gonad weights and fecundity recorded in Table 4.1 
reveals that Pacific Farms Stock (P) recorded the highest average weight o f  female brood 
fish (1110±364.7 g) followed by ARDEC Stock (A): (1060±251.00 g) and then Kumah 
Farms Stock (K): (1020±148.32 g). For the average egg weights, Stock P brood females 
had the highest (156±53.8 g) followed by Stock K (124.7±62.67 g) and then Stock A 
(90.68±39.24). The gonado-somatic indexes o f the female broodstocks were 8.69%, 
12.23% and 14.14% for stock A, stock K and stock P respectively. The average fecundity 
for Stock P, Stock K and then Stock A are 8.57 x 104 eggs, 7.06 x 104 eggs and 6.43 x 
104 eggs respectively. Stock P recorded an average total number o f  58471 eggs hatched 
representing 68.25%, followed by stock K with 53256 eggs hatched (75.48%) and stock 
A had 46644 eggs hatched (72.54%).
For the males, the highest average weight was recorded by Stock K (975.0±244.58 g) 
followed by Stock P (887.5±252.98 g) with Stock A (581.30± 151.00 g) recording the 
least. Stock P males recorded the highest gonad weight (4.17±3.75 g). This was followed 
by Stock K (2.92±2.15 g) and the least by Stock A (2.03±1.21 g). The gonado-somatic
index for stock K (0.09%), stock A (0.34%) and stock P (0.47%) were all less than 1 % ot 
the male broods fish body weight.
Upon subjecting the data to one-way analysis o f  variance (Appendix 2), it was realized 
that there were no significant relationship between the female broodstocks weights 
(F=0.14, P =0.871), egg weights (F=1.95, P =0.184), gonad weights (F=1.70, P = 0.207) 
as well as fecundity (F=0.55, P = 0.592) whereas a highly significant relationship existed 
between the male broodstocks weights (F=8.64, P =0.002).
The locality, number o f  eggs per female o f  C. gariepinus and the Authors (table 4.2) were 
compared. Mulder (1971) had the highest eggs per female: 293 - 446 (* 103) in 
Transvaal, South Africa followed by Micha (1973): 3 -  328 (x 103) in Ubangui 
River,West Africa, Nawar and Yoakim (1984): 13.9 -  164.8 (x 103) in River Nile, North 
Africa, Bruton (1979): 52 -  163 (x 103) in Lake Sibaya, South Africa, Richter (1976): 10
- 120  (x 103) and the least o f  64 - 85 (x  103) from Kumah farms, Pacific farms and 
ARDEC.
41
Table 4.1: The stock, sex, average fish weight, average gonad weight, average number o f  
eggs, hatchability and the Gonado-Somatic Index o f  C. gariepinus.
STOCK SEX TOTAL NO AVERAGE AVERAGE GONAD AVERAGE HATCHABILITY GSI
OFFISH FISHWT(g) W T(g) NO OF EGGS TOTAL NO PERCEN
OF EGGS TAGE (%)
ARDEC FEMALES 5 1060.00± 90.68± 64301± 46644 72.54 8.69
(A)
251.00 39.24 30686
MALES 8 581.30± 2.03± 0.34
151.00 1.21
FEMALES 5 1110.00± 156.00± 85672± 58471 68.25 14.14
PACIFIC
(P) 364.7 53.80 30281
MALES 8 887.50± 4.17± 0.47
252.98 3.75
KUMAH
FEMALES 5 1020.00± 124.7± 70556± 53256 75.48 12.23
(K)
148.32 62.67 33566.28
MALES 8 975.00± 2.92± 0.09
244.58 2.15
42
Table 4.2: The locality, number o f eggs per female o f C. gariepinus and the Authors
LOCALITY NO. OF EGGS PER AUTHORS
FEMALE(x 103)
Transvaal, South Africa 293 - 446 M ulder (1971)
Ubangui River,W est Africa 3 - 3 2 8 M ich a(1973)
Central and W est Africa 1 0 - 120 Richter (1976)
Lake Sibaya, South Africa 5 2 -1 6 3 Bruton (1979)
River Nile, North Africa 1 3 .9 -1 6 4 .8 Nawar and Yoakim (1984)
Kumah farms. Pacific farms 6 4 -8 5 Thesis (current study)
and ARDEC
4.1.1 Relationship between fecundity, gonad weight and fish body weight 
Stocks P, A and K according to Fig. 4.1 show a correlation o f (r = 0.809), (r = 0.434) and 
(r = 0.281) respectively between the fecundity and the fish body weight. The Stock K had 
the strongest correlation among the three followed by Stock A and then Stock K. The 
fecundity also correlates with gonad weight for the three Stocks. For Stock P (r = 0.999). 
Stock K (r = 0.975) and Stock A (r = 0.974) (Fig. 4.2).
43
KUMAH FARMS
14 v - 0.006* ■> 0.574 
12 R= 0.281
10 ♦
3 
6
4 
2 
0
700 800 900 1000 1100 1200
1300
TOTAL WEIGHT (g)
PACIFIC FARMS
±A
12
lO
8
G
A
2
O
lOOO 1200 1600
TOTAL W EIGHT <g)
ARDEC
1 A 
12 y = 53.06x4 8051. 
K 0.434
lO  
8 
6 
A 
2 
O
800 lOOO 1200
TOTAL WEIGHT <g)
Fig-4.1: Relationship between fecundity and fish body weights o f  the three stocks o f  C. 
gariepinus.
44
KUMAH FARMS
12
5X 10
‘SB
8
£
o 6
z
3 4
UJ
2
0
100 150 250
GONAD WEIGHT (g)
PACIFIC FARMS
1 4
1 2 562.6x - 2100 
R 0.009
l O
X
oo 3
<5
4
2
O
50 lOO 150 200 250
G O N A D  W E IG H T  <K)
ARDEC
14 
12
S  l O
X
-25 8
£z=> 6 ZD 
£  4
2
O
5 0  l O O  1 5 0 2 0 0
GONAD WEIGHT {g)
Fig-4.2: Relationship between fecundity and gonad weights o f the three stocks o f  C. 
gariepinus.
45
4.2 Growth Parameters
Table 4.3 shows the mean relative growth and survival o f  the three stocks o f  C. 
gariepinus. Initial mean body weight o f the stocked fry for stock A, stock K and stock P 
were 0.00300 g, 0.00325 g and 0.00313 g respectively. The highest mean weight o f  
fmgerlings at the end o f  the experiment was 3.496 g (Stock A) and lowest was 3.256 g 
(Stock K). Mean daily growth rate ranged between 0.0832 g/d for stock A and 0.0776 g/d 
for stock K. Survival rate o f  the fishes were all below 50%. It ranges between 37.23% 
(Stock A) and 40.92% (Stock P). Stock A recorded the highest Specific growth rate 
(16.81) whereas Stock K recorded the lowest (16.50).
Table 4.3: Mean Relative growth and survival o f  three stocks o f  C. gariepinus.
Parameter Stock A Stock P Stock K
Culture period(days) 42 42 42
No o f times Cultured 5 5 5
Stocking density(m2) 10 10 10
Mean Body W i(g) 0.003 ± 0.00325 ± 0.00313± 
0.00014 0.00035 0 . 0 0 0 1 0
Mean Body W 2 (g) 3.496 ± 3.304 ± 3.256 ± 
0.32020 0.29870 0.41680
Mean daily growth 0.0832 0.0787 0.0776
rate (g/d)
Survival rate (%) 37.23 39.67 40.92
Specific growth 16.81 16.54 16.50
rate(%/d)
46
4.2.1 Condition Factor
Table 4.4 shows the trends in Condition Factor o f  C. gariepinus for the three Stocks. The 
Condition Factor o f  Culture I for Stock A was 1.01. This increased to 1.18 for Culture II 
and then 1.28 for Culture III. It however decreased to 1.23 for Culture IV and then to 1.2 
for Culture V. There was an increment from 1.06 in Culture I to 1.27 in Culture II in the 
Condition Factor for Stock K. This declined to 1.25 in Culture III but increased again to 
1.34 in Culture IV and further reduced to 1.24 in Culture V. Stock K recorded an increase 
in Condition Factor from 1.03 in Culture I to 1.20 in Culture II. This declined to 1.19 in 
Culture III but increased again to 1.3 in Culture IV and declined further to 1.25 at the end 
of the study.
Table 4.4: Condition factor for Stocks A, K and P for the five culture times
CULTURE CONDITION FACTOR
STOCK A STOCK K STOCK P
I 1.01 1.06 1.03
II 1.18 1.27 1 .20
III 1.28 1.25 1.19
IV 1.23 1.34 1.3
V 1.2 1.24 1.25
MEAN 1.18 1.23 1.19
4.2.2 Body Length-Weight relationship
The Body Length-Weight relationship indicates a positive correlation between Stocks A 
(r = 0.954), P (r = 0.936) and K (r = 0.939). The growth pattern determinant which is the 
‘b ’ value for ARDEC stock (2.8629), Pacific farms stock (2.6076) and Kumah farms 
stock (2.7843) were all closer to 3 which suggests that the growth o f  the fishes was
47
isometric, a type o f growth that occurs at the same rate for all parts o f an organism so that 
its shape is consistent throughout development.
ARDEC Sample size =200
1.2 LogWt= 2.862Log SI -4.707
1 R 0.954
3  0.8
§ O.b i 
■sc
3 0.4 
0.2 
0
1.7 1.8 1.9 2 2.1
Log SI (mm )
PACIFIC FARMS Sample size =200
3: o.s 
* 0.4
K UMAH FARMS Sample size =200
L o g  W t = 2 .7 S 4 3 L o g S I  -4 .5 7 6 2  
1.2 R  =  0  93 9
& 10.8
3; 0.6
0.4
0.2
O
1.6 1.7 18  1.9
Log SI (nun) 2.1
Fig.4.3 Logarithmic Length-Weight relationships o f  ARDEC stock. Pacific stock and 
Kumah stock.
4.3 Water quality Parameters
Table 4.3 presents the mean water quality parameters taken throughout the culture period. 
All the water quality parameters recorded during the study period indicate that the 
medium o f culture was conducive for the growth o f the fish species since all were within 
the ranges reported by De Graaf and Janssen (1996). Temperature, Dissolved Oxygen and 
pH ranges from 30.88±0.75 (stock A) to 31.15±0.87 (stock P), 4.28±2.34 (stock A) to 
4.99±2.45 (stock K) and 6.18±0.15 (stock P) to 6.98±0.61 (stock A) respectively. 
Turbidity was between 11.00±4.43 (stock P) and 13.83±4.12 (stock K), Total Suspended 
Solids (20.33±8.43 to 27.83±10.12), Ammonia (0.12±0.06 to 0.43±0.37), Nitrite 
(0.01±0.02 to 0.02±0.02), Nitrate (0.05±0.04 to 0.08±0.04), Total Alkalinity (39.00±5.97 
to 46.33±9.03) and Total Hardness (33.50±5.99 to 41 ,83±8.75).
Table 4.4: Mean water quality parameters for stocks A, P and K
AVERAGE SAMPLE VALUES 
PARAMETER STOCK A STOCK K STOCK P
Temperature (°C) 30.88±0.75 30.92±0.6 31.15±0.87
Dissolved Oxygen (mg/1) 4.28±2.34 4.99±2.45 4.40±1.48
pH 6.98±0.61 6.367±0.18 6.18±0.15
Turbidity (NTU) 11.67±2.88 13.83±4.12 11.00±4.43
Total Suspended solids (mg/1) 27.83±10.12 20.33±5.32 20.33±8.43
Ammonia (NH3-N) (mg/1) 0.12±0.06 0.43±0.37 0.13±0.03
Nitrite (N 02-N ) (mg/1) 0 .01±0.02 0 .02±0.02 0.01±0.04
Nitrate (N 03-N ) (mg/1) 0.05±0.04 0.08±0.04 0.05±0.03
Total Alkalinity (mg/1) 39.00±5.97 39.75±8.43 46.33±9.03
Total Hardness (mg/1) 33.50±5.99 35.00±6.42 41.83±8.75
49
CHAPTER FIVE 
5.0 DISCUSSION
Fecundity which is the number o f  ripening eggs found in the female prior to the spawning 
act (Bagenal, 1978) varies greatly in individuals o f  one species o f  the same weight, length 
and age, but in many species it increases in proportion to the size o f fish (Lowe- 
McConnell, 1975). Fecundity in this study revealed that an average C. gariepinus brood 
female o f weight ranging from 1020 g to 1110 g from Kumah farms, Pacific farms and 
ARDEC produce an average number o f  within 6.4 x 104 to 8.5 x 104 eggs weighing 
between 90.68 g to 156.00 g. Differences in weights o f  female broodstocks (F = 0.14, P = 
0.871), egg weights (F = 1.95, P = 0.184) and fecundity (F = 0.55, P = 0.592) among the 
three stocks were not statistically significant which implies that females from these stocks 
produce approximately the same number o f eggs. The weights o f  the C. gariepinus 
broodstocks used for the artificial breeding were within 300 g and 2000 g as established 
by Olaleye (2005). The fecundity o f  the females from the three Stocks after the study fell 
within the ranges given by other Authors as indicated in Table 4.2.
The results shown in Fig. 4.1 indicate that Pacific farms stock showed a strong positive 
correlation (r = 0.809) between the fecundity and fish body weights followed by ARDEC 
stock (r = 0.434) and then Kumah farms stock (r = 0.281). This implies that an increase in 
fish body weight o f a female brood corresponds with an increase in fecundity for all the 
three stocks and this finding agrees with De Graaf et a l ,  (1995) who reported that heavier 
females yielded eggs o f higher number and weight expressed in grams.
50
Strong correlation (Fig. 4.2) between fecundity and its weight was recorded by Pacific 
farms (r = 0.999) followed by Kumah farms (r = 0.975), and then ARDEC (r = 0.974). 
This shows that an increase in gonad weight also corresponds with an increase in the 
number o f eggs in all the three stocks.
The data obtained in Table 4.1 show that the treatment with the pituitary hormone 
preparation o f the African catfish gave satisfactory hatching rates o f  eggs for Kumah 
farms (75.48%), ARDEC (72.54%) and Pacific farms (68.25%). Differences in hatching 
rates among the three stocks were not statistically significant. This is satisfactory because 
in fish reproduction under controlled conditions, attempts are made to obtain eggs o f  the 
highest weight possible and o f the best quality, and hence to produce the highest possible 
numbers of good quality hatch all year round. These hatching rates fell w ithin the 50 to 
80% hatching rates established by De G raaf et al., (1995) in well-managed hatcheries and 
this was as a result o f  the appropriate average temperature (29°C) and Dissolved Oxygen 
(4.5mg/l) recorded during incubation o f the eggs at the hatchery. De G raaf and Jansen 
(1996); Hogedoom (1980) and Viveen et al. (1985) confirmed this by stating that during 
incubation, the eggs should be well oxygenated for maximum hatching to occur. Bruton 
(1979) also buttressed on the fact that the principal requirement for successful incubation 
and hatching o f Clarias eggs is oxygenated, silt-free water at the correct temperature (19 
to 30°C) as these findings corresponds to the environmental temperature during the 
period o f natural reproduction.
51
The highest Gonado-Somatic Index o f  the female broods from the three stocks calculated 
was 14.14% and fell within the 12 to 20% reported by De G raaf and Janssen (1996). This 
proves that all the female broods used for the study were in good condition (gravid) for 
artificial reproduction o f the fish species. No significant statistical difference (F = 2.62, P 
= 0.114) however existed among all the Gonado-Somatic Indexes o f  the three stocks.
The highest final mean weight gain (3.496 g) according to Table 4.2 was observed in 
ARDEC stock and the lowest mean weight gain (3.256 g) was observed in the Kumah 
farms stock but no significant difference (F=0.66 , P = 0.533) was observed within the 
three stocks (Appendix 2). This indicates that the fries from all the three stocks grew at 
approximately the same rate. Growth is density dependent; the higher the rearing density 
of larvae, the lower their growth rate. Stocking density o f  C. gariepinus is considered the 
most important factor affecting cannibalism and aggression. It is known to have a strong 
influence on growth, survival and behaviour o f  fish (Hecht and Appelbaum, 1988). 
Micha (1976) indicated that protected ponds could be stocked within 10 to 100 fries per 1 
irf but the lower the stocking rate, the higher the growth rate o f  C. gariepinus. In this 
study however, the stocking density o f 10 fries per 1 m2 contributed to the high growth 
rate o f the three stocks o f  C. gariepinus. It also enhanced the survival rate since the fries 
had enough space to feed and also escape from their predators.
The condition factor is used to compare the “condition”, “fatness” or wellbeing o f  fish 
and it is based on the hypothesis that heavier fish o f  a particular length are in a better
52
physiological condition (Bagenal. 1978). It represents how fairly deep bodied or robust 
fishes are (Reynold, 1968). The trends in condition factor o f  the fmgerlings from all the 
three stocks after the various culture times in this study as shown in Table 4.3 was 
relatively high (above 1) and stable. No significant difference (F = l. 16, P = 0.345) existed 
among the stocks. This underscore the fact that all the fishes harvested after the study 
were in good condition.
The Length-Weight relationship (Fig. 4.3) shows a strong correlation between the length 
and weight o f all the three stocks with ARDEC stock (r = 0.954) recording the highest 
followed by Kumah farms stock (r = 0.939) and then Pacific farms stock (r = 0.936). This 
reveals that an increase in length o f C. gariepinus corresponds w ith an increase in body 
weight o f the fish species for all the three stocks. The growth pattern determinant which 
is the ‘b ’ value for ARDEC stock (2.862), Pacific farms stock (2.609) and Kumah farms 
stock (2.784) were all closer to 3 which suggests that the growth o f  the fishes was 
isometric, a type o f  growth that occurs at the same rate for all parts o f  an organism so that 
its shape is consistent throughout development.
Survival rate o f the three stocks o f  fishes (Table 4.3) were all below 50%. It ranges 
between 37.23 (Stock A) and 40.92 (Stock P). This survival though below 50%, is 
commendable since survival rate in the fish species as indicated by De G raaf et al.,
(1995) could fall as low as 5% or even 0% (100% Mortality). De G raaf et a l., (1995) 
further stressed that the successful hatching o f fish eggs and careful feeding o f larvae
53
during the early stages o f their development is essential for better survival of larvae. No 
significant difference (F=0.55, P = 0.601) existed among all the three stocks. This 
survival rate is attributed to the fertilization o f the ponds with organic manure (poultry 
droppings) prior to stocking thereby producing enough phytoplankton for the fish to 
consume since this is the most critical factor for the successful nursing o f  the catfish 
larvae during the first week after stocking o f  the fry.
According to De G raaf et al., (1995), factors influencing fry production in ponds include 
the presence o f  amphibian predators like adult frogs which can cause mortality o f  10% 
due to predation and competition for food resources. Phytoplankton levels, in the 
unprotected ponds, are reduced by the presence o f  phytophagous frog larvae, 
consequently leading to a reduction in availability o f  zooplankton, needed by the catfish 
larvae. Hogendoom, (1979), Hogendoom et al., (1976) and M icha, (1976) also realized 
that the production o f  C. gariepinus in unprotected ponds is highly variable and that the 
production in protected ponds proved to be reliable and is about 8 times higher than in 
unprotected ponds. Therefore the fencing o f  the study ponds with a mosquito net mesh 
size happas prevented predatory frogs from preying on the fry and this contributed to the 
moderate survival rates.
De Graaf and Janseen (1996) reported that differential sizes that exist in fry after stocking 
is very high after the first six weeks o f stocking and this promotes high cannibalism 
among the fry leading low survival rates. Micha, (1976) also stated that harvesting fry
54
after 3 5 ^ 0  days is considered to be optimal. The harvest o f  fry at the end o f  every six 
weeks in this study contributed to the high survival rate o f  fry since predominant sibling 
cannibalism that exist in culture ponds after six weeks o f culture as observed by Janssen 
(1985) and Hecht et al. (1988) was prevented.
All the water quality parameters recorded during the study period according to Table 4.3 
indicate that the medium o f culture was conducive for the growth o f  the fish species since 
all the physico-chemical factors were within the ranges reported by De G raaf and Janssen 
(1996). These researchers stated that the abiotic factors (physico-chemical factors) o f  the 
water environment affect the behavior, biology and growth and this determines the 
fishes’ abundance and distribution in the water.
55
CHAPTER SIX 
6.0 CONCLUSIONS AND RECOMMENDATIONS
6.1 CONCLUSIONS
The study shows that the artificial reproduction and mass rearing o f  C. gariepinus 
throughout the year is technically possible under tropical conditions by using ‘low cost 
adapted methods. An increase in both fish body weight and gonad weight o f a temale 
brood corresponds with an increase in fecundity for all the three stocks and this finding 
agrees with De G raaf et al., (1995) who reported that heavier females yielded eggs o f 
higher number and weight expressed in grams. The hatching o f  the eggs was successful 
as hatching rates for the broodstocks from all the three stocks fell within the 50 to 80% 
recommended hatching rates in well-managed hatcheries (De G raaf et al., 1995). The 
survival rate o f the fries from all the three stocks though below 50% (between 37.23% 
and 40.92%), is commendable since survival rates in the C. gariepinus fries as indicated 
by De Graaf et al., (1995) could fall as low as 5% or even 0% (100% Mortality).
All the water quality parameters recorded during the study period according to Table 4.3 
indicate that the medium o f culture was conducive for the growth o f the fish species since 
all the physico-chemical factors were within the ranges reported by De G raaf and Janssen
(1996). The growth pattern determinant which is the ‘b’ value for ARDEC stock (2.862), 
Pacific farms stock (2.609) and Kumah farms stock (2.784) were all closer to 3 which 
suggests that the growth o f the fishes was isometric, a type o f  growth that occurs at the
56
same rate for all parts o f an organism so that its shape is consistent throughout 
development.
No statistical difference was found among the mean growth rates o f  ARDEC stock (3.496 
g), Pacific farms stock (3.304 g) and Kumah farms stock (3.256 g) indicating that C. 
gariepinus grows at approximately the same rate in pond culture system regardless o f  its 
geographical location. This implies that fish farmers can reduce cost and save time and 
resources as they can rely on the nearest fish seed (fmgerlings) for stocking.
57
6.2 RECOMMENDATIONS
The main limitation on the expansion o f catfish culture in Ghana is the inadequate supply 
o f high-quality seed, especially at the right time and place, for stocking purposes. It is 
totally dependent on the government, NGOs and research institutes to produce and supply 
the fish seed. One way to overcome this constraint is to develop and promote low-input 
systems with local materials that are quite simple and easily transferable to rural fish 
farmers as done in this study.
Further studies are therefore needed to increase the viability, survival and development 
o f larvae used in farming conditions. Fries o f  C. gariepinus should therefore be cultured 
in concrete tanks and hapas fixed in ponds in order to compare the early life growth and 
survival rates o f the fish species from the three stocks to the pond culture o f  the fry 
undertaken in this study.
Stress in broodstocks should be minimal to guarantee optimal gamete quality production 
in fish. External factors such as confinement, starvation, transportation and dryness result 
into predictable pattern o f physiological changes which among other things attributed to 
weight loss and poor gamete quality o f stressed broodstock o f  C. gariepinus
58
REFERENCES
Adeyemo A. A., Ayinla O.A., and Oladosu E.A. (1992). Growth and survival o f the fry o f  
selected culturable catfish species nursed on the cultured Monia dub,a comparison 
with other sources. Nigerian Institute o f  Oceanography and Marine Research (NIOMR), 
Technical Paper 78.
Anonymous (1987). Les priorites pour la recherche aquicole en Afrique. Compte rendu 
d'un atelier a Dakar,Senegal, 1986, le Centre de Recherche pour le Developpement 
International,MR 149f, Ottawa (Canada).
Bagenal T.B. (1971). The interpretation o f  the size o f  fish eggs, the date o f  spawning and 
the production cycle. J. Fish Biol. 3:207-219.
Bagenal T.B. (1978). Aspects o f  fish fecundity. In: S.D. Gerking (Ed) Ecology o f 
Freshwater fish Production. Blackwell Scientific Publications, Oxford: 75-101.
Bard J.D.E., Kimpe P., Lizard S., Lemansson N.J., and Lentessent T.P. (1976). Hand 
book o f Tropical Culture Center Tech. Paper. Forestlier Trop. France.
Barton B.A., and Iwama, G.K. (1991). Physiological changes in fish from stress in 
aquaculture with behaviour o f  African catfish. Aquaculture International 3. 217-225.
Benau D. and Temer. C. (1980). Initiation, prolongation and reactivation o f  the motility 
of salmonid spermatozoa. Gamete Res. 3: 247-257.
Beyer J.E. (1987). On length-weight relationship. Part 1. Corresponding the mean weight 
o f a given length class. Fishbytes 5(1): 1 1 -1 3 .
Bolger T. and Connoly P.L. (1989). The selection o f  suitable indices for the measurement 
and analysis o f  fish condition. J. Fish Biol. 34: 171 -  182.
Bozkurt Y. (2006). Relationship between body condition, sperm quality parameters and 
fertilization success in Rainbow trout (Oncorhynchus mykiss). Journal o f  Animal and 
Veterinary Advances 5 (4): 284-288.
Bozkurt Y„ Secer S., Bukan N., Akcay E. and Tekin N. (2006). Relationship between 
body condition, physiological and biochemical parameters in brown trout (Salmo trutta 
fario) sperm. Pakistan Journal o f Biological Sciences. 9 (5): 9 4 0 -9 4 4 .
Bromage N.R., Jones J., Randall C., Thrush M., Davies B., Springate J., Duston J and 
Barker G. (1992). Broodstock management, fecundity, egg quality and the timing o f  egg 
production in the rainbow trout (Oncorhynchus mykiss). Aquaculture 100: 141-166 &
59
Bromage N.R., E llio, J.A., Springa.e J.R.C. and W hitehead C  (1984). The effect o f  
constant photoperiods on (he timing o f spawning in the rainbow trout. Aquaculture
43:213-223.
Broom D.M. (1998). Fish welfare and the public perception on farmed fish. In: Report 
Aquavision 98,The second Nutreco aquaculture Business conference. Stanvanger forum, 
Norway. 13-15 May. Colin E.C., and Julien, V. ed. pp. 89-91.
Bruton M.N. (1979). The food and feeding behaviour o f  Clarias gariepinus (Pisces, 
Clariidae) in lake Sibaya, South Africa, with its emphasis on its role as a predator o f 
cichlids.
C.T.F.T. (1972). Premieres directives pour l'introduction de Clarias lazera en 
pisciculture. SF/RAF/66/054: Annexe 8. Centre Technique Forestier Tropical, Nogent- 
sur-Mame, 16 pp. cages. Aquaculture 152. 67-76.
Clay D. (1977). Preliminary observations on salinity tolerance o f  Clarias lazera  from 
Israel. Bamidgeh 29(3): 102-109.
Clay D. (1979). Population biology, growth and feeding o f  the African catfish, Clarias 
gariepinus, with special reference to juveniles and their importance in fish culture. Arch. 
Hydrobiol., 87 (4): 453-482.
Clay D. (1981). Sexual maturity and fecundity o f  the African catfish (Clarias gariepinus) 
with an observation on the spawning behavior conditions. Journal o f  Zoology London 
214.21-44.
Cruz-Casallas P.E., Lombo-Rodriguez D.A. and Velasco-Santamaria Y.M. (2005). Milt 
quality and spermatozoa morphology o f captive Brycon siebenthalae (Eigenmann) 
broodstock. Aquaculture Research. 36: 682-686.
Davy F.B., Chouinard A. (1980). Induced breeding in South East Asia. The International 
Development Research Centre, Ottawa, Canada.
De Graaf G.J. (1989). La reproduction artificielle et l'alevinage de Clarias gariepinus au 
centre de production d'alevins de Loka en Cote-d’Ivoire. Rapport d'une mission effectuee 
du 28/10/1989 au 10/11/1989. Projet du developpement de la pisciculture rurale 
FAO/IVC/87/007. (unpublished).
De Graaf, G.J., Galemoni, F. and Banzoussi, B. (1995). The artificial reproduction and 
fingerling pioduction ol the African catfish Clarias gariepinus (Burchell 1822) in 
protected and unprotected ponds. Aquaculture Research 26: 233-242.
60
De Graaf G.J. and Janssen H- (1996). Artificial reproduction and pond rearing o f  the 
African catfish, Clarias gariepinus, in sub-Saharan Africa. Fisheries Technical Paper 
362. FAO, Rome.
De Leeuw R., Goos H.J.. Richter C.J.J. and Eding E.H. (1985). Pimozide-LHRHa 
induced breeding o f  the African catfish, Clarias gariepinus. Aquaculture, 44, 295-302,
Diaz L.S., Roa A., Garcia C.B., Acero A. and Navas G. (2000). Length-weight 
relationships o f  demersal fishes from the upper continental slope o ff Colombia. The 
ICLARM Quarterly 23(3): 23-25.
Eding E.H., Janssen J.A.L., Kleine Staarman G.H.J. and Richter C.J.J. (1982). Effect o f 
human chorionic gonadotropin (HCG) on maturation and ovulation o f oocytes in the 
ovary o f  the African catfish, Clarias lazera. In: Richter C.J.J., Goos H.J.th. (eds.): Proc. 
Int. Symp. Reproductive Physiology o f Fish, 2 -6  August 1982, W ageningen, the 
Netherlands. Pudoc, Wageningen, p. 195.
El Bollock A.R. (1976). Rearing o f the nile catfish, Clarias lazera  to marketable size in 
Egyptian experimental ponds. Symp. FAO/CPCA on Aquaculture in Africa. Accra, 
Ghana. CIFA Techn. Pap. 4 (1): 612-620. emphasis on the response and effects o f  
corticosteroids. Annual Review o f Fish Diseases 1, 3-4.
Food and Agriculture Organisation (FAO). (2005). Report o f  the FAO-W orldfish Centre 
on Small-Scale agriculture. Revisiting the aquaculture target group paradigm. Limbe, 
Cameroon, 23-26 March 2004. 36. http://vlib.wf.cslive.org/dbtw- 
wpd/Pdf/Staff/W F_718.pdf (Date Read: 01.04.08).
Gall G.A.E. (1974). Influence o f  size o f  eggs and age o f female on hatchability and 
growth o f rainbow trout. Calif. Fish Game 60:26-35.
Garcia C.B., Buarte J.O., Sandoval N., Von Schiller D. and Najavas P. (1989). Length- 
weight Relationships o f Demersal Fishes from the G ulf o f  Salamanca, Colombia Fishbyte 
2 1 :3 0 -3 2 .
Goos H.J., Joy K.P., De Leeuw R., Van Oordt P.G.W .J., Van Delft A.M.L. and Gielen J. 
(1987). The effect o f  luteinizing hormone-releasing hormone analogue (LHRH-a) in 
combination with different drugs with anti-dopamine and anti-serotonin properties on 
gonadotropin release and ovulation in the African catfish, Clarias gariepinus 
Aquaculture, 63, 143-156.
Graham E.F., Schmehl M.K.L., Evenson B.K. and Nelson D.J. (1978). Viability assavs 
for frozen semen. Cryobiology 15: 242-244.
Haimovici M. and Velasco G. (2000). Length-weight relationship o f  marine fishes from 
southern Brazil. The ICLARM Quarterly 23 (1): 14-16,
61
Havlor G S (1993). Controlled hatchery production o f  African catfish , Clarias 
gariepinus (Burchell): an overview. Aquaculture and Fisheries M anagement 24, - 4 5 -5 2 .
Hecht T (1996). An alternative life history approach to the nutrition and feeding o f 
Siluroidei larvae and early juveniles. Alternative life history strategies o f  catfishes. 
Aquatic Living Resources 9 (Hors Serie). 121-133.
Hecht T. and Appelbaum S. (1988). Observations on intraspecific aggression and coeval 
sibling cannibalism by larval and juvenile Clarias gariepinus (Clariidae: Pisces) under 
controlled conditions. J. Zool., Lond., 214: 21-44.
Hecht T., Uys W. (1997). Effect o f  density on the feeding and aggressive behaviour in 
juvenile African catfish, Clarias gariepinus. South African Journal o f  Science 93, 5j>7- 
541.
Hecht T., Uys, W. and Britz, P.J. (1988). The culture o f  sharptooth catfish, Clarias 
gariepinus in southern africa. South African National Scientific Programmes Report no 
153
Hogendom H., Vismans M.M. (1980): Controlled propagation o f  the African catfish 
Clarias lazera (C.and V.) II. Artificial reproduction. Aquaculture, 21, 39-53.
Hogendoom H. and Wieme R. (1976). Preliminary results concerning the culture o f  
Clarias lazera in Cameroon. Symp. FAO/CPCA on Aquaculture in Africa. Accra, Ghana. 
CIFATechn. Pap. 4 (1): 621-624.
Hogendoom H. (1979). Controlled propagation o f the African catfish, Clarias lazera 
(C&V). I. Reproductive biology and field experiments. Aquaculture, 17 (4): 323-333
Hogendoom H. (1980). Controlled propagation o f  the African catfish, Clarias lazera 
(C&V). III. Feeding and growth o f fry. Aquaculture, 21: 233-241.
Howerton R. (2001). Bets management practises for Hawaiian Aquaculture. Centre for 
Tropical and subtropical Aquaculture Publication. 148: 7-31
Huisman E.A. and Richter C.J.J. (1987). Reproduction, growth, health control and 
aquaculture potential o f the African catfish Clarias gariepinus. Aquaculture, 63, 1-14.
Huisman E.A., Skjervold H. and Richeter C.J. (1976). Aspect o f  fish culture and fish 
breeding. Miscellaneous paper 13 (1976) Loudo Whoge School-W ageningen, The 
Netherlands.
Inyang N.M. and Hettiarchichi M. (1994). Efficacy o f  human chorionic gonadotropin 
(HCG) and crude pituitary extract o f fish and frog in oocyte maturation and ovulation in
62
African catfish Clarias gariepinus Burchell 1822 and Clarias anguillaris L. 1762. Aqna- 
cult. Fish. Manag., 24, 245-258.
Iwama G.K., Pickering A.D., Sumpter J.P. and Schreck C.B. (1997)^F ish stress and 
health in Aquaculture. Society for Experimental Biology, no. 62. Cambridge, UK. 
Cambridge University Press. 287 pp.
Iwama, G.K., Vijayan M.M., Forsyth, R.B. and Ackerman, P.A. (1999). Heat shock 
proteins and physiological stress in fish. American Zoologist 39, 901-909.
Janssen J.A.L. (1985a). Elevage du poisson-chat africain Clarias lazera  (C&V) en 
Republique Centraficaine. I. Propagation artificielle. FAO projet GCD/CAF/007/NET. 
Document technique no. 20, 37 pp.
Janssen J.A.L. (1985b). Elevage du poisson-chat africain Clarias lazera  (C&V) en 
Republique Centraficaine. II. Alevinage en ecloserie. FAO projet GCD/CAF/007/NET. 
Document Technique no. 21, 31 pp.
Janssen J.A.L. (1985c). Elevage du poisson-chat africain Clarias lazera  (C&V) en 
Republique Centraficaine. III. Alevinage et grossissement en etangs. FAO projet 
GCD/CAF/007/NET. Document Technique no. 22, 41 pp.
Jocque R. (1975). Sur quelques essais de reproduction induite chez Clarias lazera  et 
Clarias senagalensis. PNUD/AVB/FAO-IVC 256, Rapp. Techn. No. 43, 17 pp.
Kaiser H., Weyl, O. and Hecht, T. (1995). The effect o f  stocking density on growth, 
survival and agonistic.
Khwuanjai Hengsawat W. and Pomchai Jaruratjamom.F.J. (1997). The effect o f  stocking 
density on yield, growth and mortality o f  African catfish (Clarias gariepinus Burchell 
1822) cultured in cages. Aquaculture 152, 67-76
King R.P. (1996a). Length-weight relationships o f  Nigeria Freshwater fishes Naga 
ICLARIM Q 19(3): 4 9 - 5 2 .
King R.P. (1996b). Length-weight relationship o f  Nigerian Coastal water fishes. 
Fishbyte, 19(4): 53 -  58.
Kjorsvik E„ Mangor-Jensen A. and Holmefjord I. (1990). Egg quality in fishes In- 
Blaxter, J.H.S., Southward, A.J. (Eds.), Adv. Mar. Biol., 26: 71-113.
Koufil .(., Hamackova J. and Barth T. (1992). Induction o f  ovulation in African catfish 
Clarias gariepinus using GnRH analogue, dopaminergic inhibitor o f  isophoxythepin and
S C 5 & T S “  C0" f  ' ZOOl° SiCal S W i° "  ° f  Slovac Academy o f
63
Krause J., Jean-Guy J. and Brown D. (1998). Body length variation w i t h i n  multi-species 
fish shoals: the effects o f shoal size and number o f  species. Oceologia, 114: b/ - / z .
Kuczyriski M., Mis J. and Szumiec J. (1999). Chow suma afrykanskiego w obiegach 
recyrkulowanych. Osrodek Doradztwa Rolniczego w Bielsku-Bialej, 1-2.
Le Cren E.D. (1951). The lenght-weight relationship and seasonal cycle in gonad weight 
and conditions in the perch Perea fluviatilis. Jo u rn a l o f A nim al Ecology, London, 20 
(2): 201-219.
Legendre M. and Jalabert B. (1988). Physiologie de la reproduction. In: C. Leveque, 
M.N. Bruton and G.W. Ssentongo (eds), Biology and Ecology o f  African Freshwater 
Fishes, Editions de L ’ORSTOM, Paris.
Lowe-McConnel R.H. (1975). Fish Communities in Tropical Fresh Waters. London.
Luff R.M and Bailey G.N. (2000). Analysis o f  size changes and incremental growth 
structures in African catfish Synodontis schall (schall) from Tell el-Amama, Middle 
Eygpt. J. Arch. Sci., 27: 821-835.
Madu C.T., M ohammed S., Mezie A., Issa J. and Ita E.O. (1993). Comparative growth, 
survival and morphometries o f  Clarias gariepinus, Heterobranchus bidorsalis and their 
hybrid fmgerlings. pp. 56-61. In: Annual Report, new Bussa, N iger State (Nigeria): 
National Institute for Freshwater Fisheries Research.
Micha J.C. (1973). Etude des populations piscicoles de l'Ubangui et tentative de selection 
et d'adaptation de quelques especes a l'etang de pisciculture. Centre Technique Forestiere 
Tropical, Nogent sur Marne, 100 pp.
Micha J.C. (1975). Synthese des essais de reproduction, d'alevinage, et de production 
chex un silure Africaine, Clarias Lazera. Bull.Fr.Pisc.,276:78-87
Micha J.C. (1976). Synthese des essais de reproduction, d’alevinage et de production chez 
un silure Africain: Clarias lazera Val. Symp. FAO/CPCA on Aquaculture in Africa 
Accra, Ghana. CIFA Techn. Pap. 4 (1): 450-473.
Moussa T.A. (1956^ Morphology o f the accessory air breathing organs o f  the teleost 
Clarias lazera (C and V). J. Morph., 98:125-160.
64
Mulder P. F. S. (1971). On Ekologiese studie van die h e n g e lv is fa u n a m  die 
Vaalriviersistem met spesiale verwysingna Barbus kimber eyensis,
Thompson, M.Sc. thesis, Rand Afrikaans University, South Africa: in Bruton,
Nawar G. and Yoakim, E. G. (1984). A study on the stream o f  the Nile catfish C /ar/as 
lazera Valenciennes in Cuvier and Valenciennes 1840, Annals and M agazine o atura 
History 5 (13): 385-389.
Olaleye V.F. (2005). A review o f reproduction and gamete management in the African 
catfish, C. gariepinus (Burchell). Ife J. Sci. 7(1): 63-70.
Ovredal J.T and Totland B. (2002). The scantrol fish meter for recording fish length, 
weight and biological data. Fish. Res., 55: 325-328.
Owiti D.O. and Dadzie S. (1989). Maturity, fecundity and the effect o f  reduced rainfall 
on the spawning rhythm o f a siluroid catfish, Clarias mossambicus (Peters). Aquaculture 
and fisheries management, 20: 355-368.
Owodeinde F.G., Ogunlaru K.A., Denloye A.A.B., Irunokhai F.O. and Hammed A.M ., 
(2004). The effects o f  delayed initialfeeding on the survival o f  artificially-reared Clarias 
gariepinus fry. Journal o f  Tropical Biosciences 4: 19-22.
Pauly D. (1985). Fish population dynamic in tropical waters: a manual for use with 
programmable calculators. Naga, ICLARM Qtr., 5-95.
Pham A. (1975). Donnees sur la reproduction en masse d'alevins de Clarias lazera Val. 
(Pisces, Clariidae). Ann Univ. Abidjan, Ser. E, 8 (1):139-145. PhD thesis W ageningen 
University. 160 p.physiological stress in fish. American Zoologist 39, 901-909. 
physiological stress in fish. American Zoologist 39, 901-909.
Pickering A.D., Pottinger T.G. and Balm P.H.M. (1995). Sexual maturity modifies 
responsiveness o f  the pituitary -  inter renal axis to stress in male rainbow trout. Gen. 
Comp. Endo. 98: 311-320.
PiUay T.V.R. (1990). Aquaculture principles and practices. Fishing News Books, 
London.
Quick A.J.R. and Bruton M.N. (1984). Age and growth o f Clarias gariepinus (Pisces: 
Clariidae) in the P.K. Le Roux Dam, South Africa. S. Afr. J. Zool., 19(1): 37-45.
Rana K.J. (1990). The influence o f  maternal age and delayed initial feeding on the 
survival and growth o f previously unfed Oreochromis niloticus (L) and O mossambicus
fry. Aquaculture 91: 295-300. Research 26: 233-242.
65
Reynold T.D. (1968). The biology o f the clupeids in the New Volta. In. Man made 
Lakes. The Accra Symposium. Ghana University Press, Accra.
Richter C.J.J., Eding E.H. and Bloem A.J. (1985). 17 a-Hydroxyprogesterone induced 
breeding o f the African catfish, Clarias gariepinus (Burchell), wit out priming 
gonadotropin. Aquaculture, 44, 285-293.
Richter C.J.J., Eding E.H., Goos H.J., De Leeuw R., Scoot A.P. and Van der Oordt 
P.G.W.J. (1987). The effect of pimozyde/LHRH-a and 17a-hydroxyprogesterone on 
plasma steroid levels and ovulation in the African catfish, Clarias gariepinus. 
Aquaculture, 63, 157-168.
Richter C. J. J. (1979). The African mudfish, Clarias lazera (C and V), a new possibility 
for fish culture in tropical regions? Misc. Pap. Landbouwhogeschool, W ageningen. 13. 
51-70: in Bruton, 1976.
Ricker W.E. (1975). Computation and Interpretation o f biological statistics o f fish 
populations. Bull. Fish. Res. Board Can., (23) Suppl. 1, Vol. 2: 519-29.
Ruane N.M. (2002). Stress and stress disorders in a teleost fish, the common carp 
Cyprinus carpio L.
Seegers L. (1996). Fishes o f the Lake Rukwa drainage. Annales Sciences Zoologiques 
278. Musee Royal de L ’Afrique Centrale, Tervuren, Belgique silure Africaine, Clarias 
Lazera. Bull.Fr.Pisc.,276:78-87.
Skelton, H.P., (1993). A complete guide to the freshwater fishes o f  Southern Africa. 
Southern Book Publishers Ltd., Johannersberg, 388 pp.
Springate J.R.C., Bromage, N.R. and Cumaranatunga P.R.T. (1985). The effects o f  
different ration on fecundity and egg quality in the rainbow trout (Salmo gairdneri) pp. 
371-391. Ed: C.B. Cowey, A.M. Mackie, J.G. Bell (1997). Nutrition and Feeding in 
Fish. Academic Press, London, UK. Statistica PL for Windows, Version 5 Stat Soft, 
Polska.
Temer C. (1986). Evaluation o f salmonid sperm motility for cryopreservation. The 
Proggressive Fish Culturist 48: 230-232.
Teugels G.G., (1982a). Preliminary results o f a morphological study o f  five nominal 
species o f  the subgenus Clarias (Pisces; Clariidae). J. Nat. Hist., 16 (3): 439-464
Teugels G G. (1982b). Preliminary data o f a systematic outline o f  the African species o f 
the genus Clarias (Pisces; Clariidae). Rev. Zool. afr., 96 (4): 731 -748
66
Teugels G G  (1984). The nomenclature o f African Clarias species used in aquaculture. 
Aquaculture, 38: 373.374.Tnms. Zool. Soc. London. 35:47-114. The importance o f  fish 
in our diet and life, (www.blurtit.com).
Verreth J., Eding E.H., Rao G.R.M., Huskens F. and Segner H. (1991). Feeding practices, 
growth and nutritional physiology in larvae o f  the catfish Clarias gariepinus and C lar ias 
batrachus. In: Lavens, P., Sorgeloos, P., Jaspers, E. and Ollevier, F. (Editors), 
Proceedings International Symposium on Fish and Crustacean Larviculture (in press).
Viveen W.J.A.R., Richter C.J.J., Van-Oordt P.G.W. Janssen J.A.L and Huisman E.A. 
(1985). Practical manual o f  the culture o f  African catfish, Clarias gariepinus'. Directorate 
General International Cooperation for the Ministry o f  Foreign Affairs. The Hague, the 
Netherlands.
Volckaert F.A. and Agnese J.F. (1996). Evolutionary and population genetics o f  the 
Siluroidei. Aquatic Living Resources 9 (Hors Serie):81-92.
Wendelaar-Bonga A.E. (1997). The stress response in fish. Physiological reviews 77, 
591-625.
Willoughby N. G. and Tweddle D. (1978). The ecology o f the catfish Clarias gariepinus 
and Clarias ngamensis in the Shire Valley, Malawi, Journal o f  Zoology (London) 186: 
507-534,.
Wootton R. J. (1992). Ecology o f  Teleost Fishes, Chapman and Hall, pp.404.
Woynarovich E. and Horvath L. (1980). The artificial propagation o f  warm-water fin 
fishes: A manual for extension. FAO Fish. Techn. Paper 201. yield, growth and mortality 
of African catfish (Clarias gariepinus Burchell 1822) cultured in
Young M.J.A. and Fast A.W. (1990). Temperature and photoperiod effects on ovarian
—  in the Chinese catfish (Clarias fuscus). Journal o f  Aquaculture in the Tropics
67
APPENDICES:
Appendix 1: Six chronological stages seen within the development o f  C. gariepinus 
oocyte (Owiti and Dadzie, 1989).
Stage 1, Immature virgin:
Macroscopic description: The ovary is colourless to translucent brown, lanceolate and 
lobular in appearance, occupying the posterior quarter o f  the body cavity. In fish larger 
than 10 cm the ovary can be distinguished from the testis due to its smoothness in 
contrast to serrated edges o f  the testis.
Histological description: Pre-vitello genesis stage or primary oocytes. The oocyte are 
small (7-10 micron) and contain no yolk. The number o f  primary oocytes increases 
through mitotic division.
Stage 2, Developing virgin:
Macroscopic description: The ovary is translucent, brown in colour and occupies about 
one third o f the length o f the peritoneal cavity. Individual oocytes are visible with the 
naked eye as tiny specks.
Histological description; Pre-vitello genesis stage or primary oocytes. The oocyte are 
small (7-10 micron) and do not yet contain yolk. The number o f  primary oocytes 
increases through mitotic division and at the end o f  this stage the oocyte increase its size 
to approximately 200 micron. Vitellogenesis is the process o f  yolk formation.
68
Stage 3, Ripening:
Macroscopic description; The ovary is opaque, brownish-green in colour, occupying abut 
one half the length o f  the ventral cavity. Eggs are visible as yellowish j_n;en or bro 
yellow granules and blood capillaries visible around the ovary.
Histological description; Endogenous vitello genesis stage. W ithin this stage the yolk ot 
the oocyte (the future reserve/feed for the hatched larvae) is formed. The origin ot the 
yolk in this stage is the oocyte itself.
Stage 4, maturing or ripe:
Macroscopic description: The ovary is large, opaque, browngreen in colour. The eggs are 
yolk laden and clearly visible to the naked eye. Ovary occupies four fifth o f  the peritoneal 
cavity. A highly developed capillary net work is visible. Eggs ooze out freely with 
pressure on the belly.
Histological description; Exogenous vitello genesis. The oocyte increases to its final size 
of 1000-1200 micron (1-1.2 mm). During this phase yolk formation in the oocyte 
increases and the origin o f the proteins needed for this process is outside the oocyte (the 
liver). A large nucleus (0.2 mm) is clearly visible a little outside o f  the centre o f  the 
oocyte. The oocytes in this stage are also called "ripe eggs". They remain in this stage 
until environmental factors (rainfall and water level rise or a hormonal injection) 
stimulate their ovulation.
69
Stage 5, Running or spawning:
Macroscopic description: The eggs are translucent, flat, with cytoplasm concentrated at 
the animal pole and visible as a reddish brown spherical cap (figure 10). This aspect 
quite distinct from the round eggs present in the ovaries before 
reproduction/hypophysation
Stage 6, Spent:
Macroscopic description: The ovary is flaccid, flabby and bloodshot with thick whitish 
tough walls. The genital aperture o f the female looks inflamed. Some translucent and 
opaque (residual) eggs visible to the naked eye. The development from stage 1 to stage 4 
is related to temperature (and o f  course age when first maturation is considered). The 
development from stage 4 to stage 5
is triggered by environmental stimuli or can be provoked by hormonal injections. This 
development process will take place once the water temperature is 20-22 °C or higher. 
After ovulation o f the ripe eggs" the majority o f the oocytes found in the ovary consists 
again o f stage 1 oocytes, the cycle is repeated and after approximately six weeks a new 
batch of "ripe eggs" is ready for ovulation.
70
Appendix 2: Analysis o f Variance (ANOVA) o f Reproductive and Growth Performance 
at 95% confidence level.
Parameter F P-value Remarks
Fish Weight (Female Brood) 0.14 0.871 > 0.05, not significant
Fish Weight (Male Brood) 8.64 0.002 highly significant
Male gonad weight 1.70 0.207 > 0.05, not significant
Egg weight 1.95 0.184 > 0.05, not significant
Fecundity 0.55 0.592 > 0.05, not significant
GSI 2.62 0.114 > 0.05, not significant
Weight gain 0.66 0.533 > 0.05, not significant
Specific Growth Rate 1.95 0.163 > 0.05, not significant
Condition factor 1.16 0.345 > 0.05, not significant
Survival rate 0.55 0.601 > 0.05, not significant