MORPHOMETRIC CHARACTERISTICS OF THE JUVENILE NORTH AFRICAN CATFISH CLARIAS GARIEPINUS (BURCHELL, 1822) FROM THE HEATED WATER AQUACULTURE

Beata Więcaszek, Stanisław Krzykawski, Artur Antoszek, Joanna Kosik, Paweł Serwotka
Department of Fish Systematics, West Pomeranian University of Technology, Szczecin, Poland

ABSTRACT

North African catfish C. gariepinus is of great commercial importance both in fisheries and aquaculture. In Poland it has been introduced for aquaculture purposes in early 90.  However this is the first detailed description of the north African catfish population cultured under condition of the heated water aquaculture in Poland. The objective of the present work was to fill a gap in the knowledge on the African catfish by examining in detail the morphometric characters.The material used to study meristic and morphometric characters consisted of 100 individuals collected from the rearing ponds, measuring from 9.0 to 26.0 cm in total length (TL), the mean length being 18.1 cm. Taxonomic analysis involved 15 morphometric and 7 meristic characters, including as well the relationships among them.

Key words: aquaculture, Clarias gariepinus, gill rakers, morphometrics, occipital process.

INTRODUCTION

The highest diversity of genus Clarias Scopoli, 1777, the most speciose genus with 32 species, is found on the African continent [16]. Some species, like Clarias gariepinus (Burchell, 1822) are of great commercial importance both in fisheries and aquaculture [27]. C. gariepinus is the native species of Africa and has drawn attention of aquaculturists because of its biological attributes that include faster growth rate, resistance to diseases and possibility of high stocking density [15]. It has an almost Panafrican distribution (absent from Maghreb, the upper and lower Guinea, the Cape province, probably Nogal province), and also naturally occurs in Minor Asia: Jordan, Israel, Lebanon, Syria and southern Turkey [28]. This species has been introduced in Europe, America and south-east Asia for aquaculture purposes [1].

Aquaculture of this species is practiced in twenty countries with the total production of 3703 mt, out of which 94.7% is reported from four countries, the bulk of which is coming from the Netherlands [4]. In Europe it was introduced in 1974 first in Cyprus and later in the Czech Republic, Slovakia and the Netherlands [8]. In Poland C. gariepinus has been introduced in early 90. [5]. However the literature on its taxonomic status is still very scarce and the morphological characteristics of the species from the Polish aquaculture is lacking.

The morphometric characteristics of this species is significant, because (especially in the Asian aquaculture) the hybrids of C. gariepinus with other species are cultured. Moreover, the closely related species like C. anguillaris or C. macrocephalus are cultured at a large scale, thus the detailed characteristics of the morphometric features may be needed to distinguish the different species and hybrids within the Clarias genus. Both hybrids of C. gariepinus and other species of Clarias in the close future are likely to be cultured in Europe and Poland, too.

Clarias gariepinus is not only a very important aquaculture species in Poland, but it is also included in the checklist of non-native fishes that occur in the fresh waters of Poland [21], with the year of introduction 1990 and the reason of introduction – aquaculture [2]. Therefore the north African catfish can occur in the environment in Poland as escapees from the aquaculture.

The objective of the present work was to fill a gap in the knowledge on the north African catfish by examining in detail morphometric and meristic characters of C. gariepinus under the fishculture condition in the heated waters.

MATERIAL AND METHODS

The material used to study meristic and morphometric characters consisted of 100 frozen specimens of C. gariepinus collected from the rearing ponds (4.5x3.2x0.8m) in the Gołysz Institute of Ichthyobiology and Aquaculture (a research institution of the Polish Academy of Sciences) in 1998. The material was the juveniles of C. gariepinus, obtained from the own spawners imported from the Netherlands. Ovulation in females took place after the Ovopel injection. The water temperature in the ponds ranged from 25 to 27°C, and fish were fed by the dry pelleted feed (51.5% of proteins and 7% of lipids).

The length of specimens examined covered the range from 9.7 to 25.4 cm in total length (TL), the mean length being 18.10 cm. For the morphometric studies four 4.0 cm wide length classes were established: 9.7–13.6, 13.7–17.6, 17.7–21.6 and 21.7–25.6 cm. The weight was established with the accuracy of 0.1 g.

Taxonomic analysis involved 15 morphometric characters and 7 meristic ones. Symbols and characters of morphometric characters, as well as results of the study are presented in Table 1. The following morphometric characters were studied: total (TL) and standard (SL) lengths, dorsal head length (= the distance from tip of snout to end of occipital process) (HL) and interorbital length (IO), occipital process length (OPL) and width (OPW), premaxillary (PMW) and vomerine (VMW) toothplate width, barbel length (BL), prepectoral (PPEL), prepelvic (PPL), predorsal (PDL) and preanal (PAL) lengths, dorsal- (DFL) and anal- (AFL) fins base length. Measurements were taken according to Agnese et al. [1], slightly modified (Fig. 1). These characters were selected on their diagnostic value as demonstrated by Teugels [27]. Additionally, the length to the width of occipital process ratio was established, as well as the distance from the dorsal fin to the occipital process (OP-D). Measurements were taken with an aid of electronic caliper and microscope Nikon SMZ 1000, coupled with the Lucia Measurement System, with the accuracy of 1 mm.n The lengths were converted to per cent fish standard length (SL) and fish head characters to per cent dorsal head length (HL).

Table 1. Metric characters of the sample studied, converted to per cent fish standard length and fish head characters to per cent dorsal head length

Character and its symbol	Range	Mean±standard error	Standard deviation SD	Coefficient of variation CV
% of  SL
TL	113.5–125.7	116.01 ± 0.10	1.41	1.22
HL	28.2–35.8	31.00 ± 0.10	1.01	3.32
IOW	10.2–12.6	11.51 ± 0.10	0.51	4.53
OPW	4.7–7.3	6.11 ± 0.01	0.52	7.41
OPL	3.6–5.6	4.61 ± 0.01	0.41	9.71
PMW	7.3–10.0	8.80 ± 0.10	0.51	6.11
VMW	6.0–8.2	7.31 ± 0.011	0.52	6.31
BL	29.6–43.0	34.71 ± 0.30	2.92	8.31
OP-D	4.84–9.33	6.58 ± 0.01	1.71	26.00
PPEL	22.1–26.6	24.40 ± 0.10	0.91	3.81
PDL	25.3–38.3	36.10 ± 0.10	1.41	4.02
PPL	43.8–54.2	47.51 ± 0.10	1.41	3.02
PAL	51.9–62.6	57.11 ± 0.20	1.52	2.71
DFL	58.7–65.1	61.90 ± 0.10	1.32	2.21
AFL	38.4–46.3	42.90 ± 0.10	1.41	3.41
% of HL
IOW	32.4–41.2	37.21 ± 0.20	1.72	4.51
OP-D	16.3–24.3	20.11 ± 0.10	1.61	21.90
OPW	14.7–24.0	19.70 ± 0.20	1.51	7.80
OPL	11.8–19.4	14.70 ± 0.10	1.51	10.01
PMW	23.5–32.7	28.40 ± 0.22	1.72	5.90
VMW	17.6–26.9	23.41 ± 0.10	1.53	6.30
BL	93.8–142.3	111.81 ± 1.00	9.61	8.61

Fig. 1. Measurement design – based on Agnese et al. 1997, slightly modified (symbols are described in Table 1)

The following meristic characters were determined: counts of branched and unbranched rays in the dorsal, anal, pectoral and pelvic fins, count of gill-rakers on the first gill arch and the vertebral count. Total vertebral number included the four anteriormost vertebrae, associated with the Weberian apparatus, and the hypural fan centrum. The number of vertebrae was established after the fish dissection.

Meristic characters and their symbols (together with the results of study) are shown in Table 7. Symbols of meristic characters followed those of Holčik [10].

All the data were subjected to statistical treatment involving standard deviation (SD), mean (M), standard error of the mean (m) and coefficient of variation (CV). After Ruszczyc [23] as variable were regarded those characters of the coefficient of variation ≥ 8%. The most frequent counts (MFC) are reported for meristic characteristics as well.

Analysis of regression, the Pearson coefficient of correlation R and coefficient of determination R² were calculated in order to find a relationship among the standard (=body) length and all morphometric characters, among the standard length and the ratio: length (OPL)/width (OPW) of the occipital process, as well as the dorsal head length and morphometric characters pertinent to the head, and the total length and all meristic characters. The Student's t test was applied to test the null hypothesis that the correlation coefficient did not significantly differ from zero.

RESULTS

The most stable morphometric characters expressed as per cent of standard length in the whole sample (Table 1) were the total length, the preanal length and dorsal-fin base length, whereas the most variable characters were those pertinent with the head: the length of occipital process and the maxillary barbel (9.7 and 8.3% of CV, respectively). The characters mentioned were also the most variable when expressed as the dorsal head length proportion. The dorsal and anal fins of the north African catfish posses long bases, constituting to 61.9 and 42.9% of standard length, respectively. Both characters are stable, in the whole sample and in the particular length classes (Table 1 and 2). The maxillary barbel is usually longer than dorsal head length (111.8% of head length) both in the whole sample and in the particular length classes as well (Table 1 and 3).

Table 2. Metric characters in % of standard length and weight [g] of C. gariepinus in length classes

	Length class TL [cm]
	9.7–13.6 (n=31)					13.7–17.6 (n=11)					17.7–21.6 (n=32)					21.7–25.6 (n=26)
	M	±	m	SD	CV	M	±	m	SD	CV	M	±	m	SD	CV	M	±	m	SD	CV
TL	116.09	±	0.20	1.09	0.94	115.79	±	0.32	1.05	0.91	116.07	±	0.16	0.93	0.8	115.82	±	0.44	2.23	1.93
IOW	11.68	±	0.11	0.59	5.06	11.60	±	0.18	0.6	5.20	11.41	±	0.07	0.4	3.51	11.48	±	0.1	0.5	4.36
OPL	4.8	±	0.09	0.48	9.91	4.58	±	0.13	0.44	9.57	4.37	±	0.06	0.36	8.15	4.53	±	0.08	0.39	8.5
OPW	5.98	±	0.09	0.48	8.07	5.89	±	0.12	0.39	6.61	6.10	±	0.08	0.47	7.67	6.32	±	0.07	0.33	5.26
PMW	8.68	±	0.11	0.61	7.08	8.66	±	0.16	0.51	5.95	9.00	±	0.09	0.51	5.72	8.81	±	0.08	0.41	4.64
VMW	7.17	±	0.10	0.57	7.99	7.17	±	0.15	0.49	6.90	7.39	±	0.07	0.4	5.37	7.25	±	0.06	0.33	4.54
OP-D	7.47	±	0.08	0.46	6.19	–			–	–	–		–	–	–	5.76	±	0.17	0.86	14.99
HL	30.98	±	0.24	1.33	4.31	30.51	±	0.26	0.87	2.85	31.16	±	0.15	0.84	2.68	31.19	±	0.15	0.77	2.47
PDL	35.99	±	0.40	2.24	6.21	36.15	±	0.28	0.93	2.58	36.31	±	0.16	0.92	2.54	36.14	±	0.17	0.87	2.39
PAL	57.23	±	0.23	1.29	2.25	57.94	±	0.41	1.35	2.33	57.46	±	0.29	1.64	2.86	56.34	±	0.28	1.43	2.53
PPL	47.78	±	0.29	1.61	3.36	47.76	±	0.38	1.25	2.62	47.33	±	0.27	1.52	3.21	47.15	±	0.23	1.16	2.45
PPEL	24.37	±	0.20	1.11	4.55	24.05	±	0.31	1.02	4.22	24.32	±	0.14	0.78	3.21	24.53	±	0.16	0.83	3.38
DFL	62.28	±	0.25	1.38	2.22	62.34	±	0.28	0.92	1.47	61.91	±	0.24	1.34	2.16	61.4	±	0.26	1.34	2.18
AFL	43.25	±	0.20	1.11	2.57	42.13	±	0.6	1.98	4.70	42.81	±	0.27	1.53	3.58	42.78	±	0.27	1.36	3.18
BL	37.12	±	0.45	2.51	6.76	35.15	±	0.83	2.74	7.80	33.85	±	0.41	2.3	6.6	32.62	±	0.33	1.66	5.1
W [g]	13.32	±	0.74	4.13	30.97	24.55	±	1.70	5.63	22.93	58.88	±	1.74	9.85	16.74	85.96	±	2.65	13.53	15.73

Table 3. Metric characters of the head of C. gariepinus in % of dorsal head length  in length classes

	Length class TL [cm]
	9.7–13.6 (n=31)					13.7–17.6 (n=11)					17.7–21.6 (n=32)					21.7–25.6 (n=26)
	M	±	m	SD	CV	M	±	m	SD	CV	M	±	m	SD	CV	M	±	m	SD	CV
IOW	37.74	±	0.39	2.17	5.76	38	±	0.45	1.49	3.91	36.61	±	0.21	1.16	3.18	36.82	±	0.26	1.34	3.65
OPL	15.52	±	0.31	1.71	11.03	15.01	±	0.4	1.33	8.86	14.02	±	0.19	1.07	7.63	14.54	±	0.24	1.21	8.29
OPW	19.34	±	0.32	1.75	9.07	19.32	±	0.46	1.54	7.95	19.6	±	0.29	1.64	8.35	20.27	±	0.18	0.94	4.64
PMW	28.05	±	0.36	1.98	7.07	28.37	±	0.41	1.35	4.75	28.89	±	0.31	1.78	6.17	28.25	±	0.23	1.19	4.21
VMW	23.16	±	0.35	1.94	8.39	23.49	±	0.38	1.26	5.35	23.72	±	0.24	1.35	5.68	23.26	±	0.18	0.92	3.95
OP-D	24.18	±	0.35	1.96	8.11	–			–	–	–		–	–	–	18.48	±	0.54	2.77	14.99
BL	119.98	±	1.59	8.85	7.37	115.17	±	2.45	8.12	7.05	108.66	±	1.27	7.2	6.62	104.6	±	1.02	5.19	4.96

The average weight of the north African catfish specimens ranged from 13.32 g to 85.96 g in the consecutive length classes of fish. This character is very variable in all length classes, with the coefficient of variation CV covered the range from 15.73 to 30.97% (Table 2).

When analyzing the mean values of particular characters in the length classes, it is seen that in case of two characters: the distance from the dorsal fin to occipital process and length barbel, the averages decreased in the successive length classes, while in next three: occipital process length as well as length of the dorsal and anal fin bases, they decreased but very slightly. On the opposite, the averages of dorsal head length and predorsal distance slightly increases in the successive length classes (Table 2). Among the head characters expressed in hundredths of head length, the maxillary barbel mean length and the distance from the dorsal fin to occipital process decreased markedly in length classes, whereas the occipital process changed its shape, as its length decreased and the width increased slightly (Table 4) with the fish total length.

Table 4. Relationship between the standard fish length and particular metric characters of C. gariepinus studied (P<0.05)

Character	Regression equation f(x)	R2	R
TL	1.15x+0.12	1.00	1.00*
IOW	0.11x+0.04	0.97	0.98*
OPL	0.04x+0.06	0.87	0.93*
OPW	0.07x-0.09	0.94	0.97*
PMW	0.09x-0.05	0.95	0.97*
VMW	0.08x-0.04	0.95	0.97*
HL	0.32x-0.08	0.99	0.99*
PDL	0.36x-0.04	0.98	0.99*
PAL	0.56x+0.19	0.99	0.99*
PPL	0.46x+0.15	0.99	0.99*
PPEL	0.25x-0.05	0.98	0.99*
DFL	0.6x+0.22	0.99	0.99*
AFL	0.42x+0.12	0.98	0.99*
BL	0.28x+1.03	0.92	0.96*

* Coefficient of correlation is statistically significant.

Table 5. Relationship between the dorsal length and particular head characters of C. gariepinus studied (P<0.05)

Character	Regression equation f(x)	R2	R
IOW	0.36x+0.07	0.98	0.99*
OPL	0.13x+0.07	0.87	0.93*
OPW	0.21x-0.06	0.93	0.96*
PMW	0.29x-0.02	0.96	0.98*
VMW	0.24x-0.01	0.95	0.97*
BL	0.87x+1.1	0.93	0.96*

* Coefficient of correlation is statistically significant.

The relationships of morphometric characters with the standard length and with the dorsal head length of the fish sample studied (Table 4 and 5) are significant statistically. All coefficients of correlation presented in these tables are positive and statistically significant at p < 0.05, and their values usually are close to 1. These relationships are straight-linear, described by the equations demonstrated in Tables 4 and 5. Particular parts of fish body increase in relation to its length (standard or head) steadily and some of them nearly directly proportionally (the interorbital length, occipital process length, premaxillary and vomerine toothplate width, predorsal and preanal lengths. In the case of head characters all of them increase them nearly directly proportionally, except the barbel length (Table 5).

The ratio of occipital process length to the occipital process width is a variable character, except the last length class: 21.7–25.6 cm of TL (Table 6). The occipital process width is bigger than its length; the average of the occipital process length to its width ratio diminished slightly in the successive length classes of fish, from 0.80 to 0.72, being slightly however statistically significantly correlated with the catfish standard length (Table 6).

Table 6. Occipital process length to the occipital process width ratio (%) in the sample of the C. gariepinus studied in length classes (P<0.05)

Length class TL [cm]

9.7–13.6 (n=31)

13.7–17.6 (n=11)

17.7–21.6 (n=32)

21.7–25.6 (n=26)

M

±

M

SD

CV (%)

M

±

m

SD

CV (%)

M

±

m

SD

CV (%)

M

±

m

SD

CV (%)

OPL/OPW

0.80

±

0.01

0.08

9.64

0.78

±

0.03

0.09

11.74

0.72

±

0.01

0.06

8.88

0.72

±

0.0 1

0.04

5.86

Coefficient of correlation R

0.44*

Coefficient of determination R2

0.19

Regression equation f(x)

y=0.001x+0.81

* – Coefficient of correlation is statistically significant.

Meristic characters of the north African catfish examined are summarised in Table 7. Values of coefficient of variation are relatively low, indicating a low meristic variability features. Except for the unbranched ray count in the pectoral fin, which was stable (Pu = 1), the vertebral count was the most stable character in the sample studied, while the branched ray counts in the pectoral fin and gill rakers on the first branchial arch were the most variable meristics. In the first two length classes of C. gariepinus (9.7 to 17.6 cm of total length), the most frequent counts of gill rakers ranged from 28 to 31, while in the next two length classes (17.7–25.6 cm of length), MFC covered the range from 40 to 50. Averages of the gill rakers count in the consecutive length classes are presented in Table 7. As can be seen, their values increase markedly with the catfish length. In all length classes this character appeared to be very variable with the coefficient of variability CV reached 24.3%.

Table 7. Meristic characters of the north African catfish C. gariepinus studied

Character and its symbol

Range

M±m

MFC

SD

CV (%)

Db – branched rays count in D fin

55–75

67.42±0.35

60–73

3.5

5.2

Vb - branched rays count in V fin

4–6

5.85±0.04

5–6

0.4

6.6

Pb – branched rays count in P fin

6–11

8.86±0.07

8–9

0.7

8.3

Pu – unbranched ray count in P fin

1

1±0.00

1

0.0

0.0

Ab – branched ray count in A fin

36–59

51.29±0.37

50–55

3.7

7.1

GR – gill rakers count on the first branchial arch

24–55

37.92±0.74

28–31; 40-50

7.4

19.5

vt. – vertebral count

52–64

58.37±0.22

58–60

2.2

3.7

Gill rakers count in length class TL [cm]

9.7–13.6 (n=31)

13.7–17.6 (n=11)

17.7–21.6 (n=32)

21.7–25.6 (n=26)

M

±

m

SD

CV (%)

M

±

m

SD

CV (%)

M

±

m

SD

CV (%)

M

±

m

SD

CV (%)

30.39

±

1.33

7.31

24.3

31.17

±

2.23

7.31

23.9

42.13

±

1.31

7.39

17.5

44.78

±

0.81

4.14

9.23

* Coefficient of correlation is statistically significant.

No statistically significant correlation was found among the meristic characters and the fish length, except the gill rakers on the first gill arch (Table 8), which are strong correlated with the total length of the fish examined. The regression equation concerns the total length-gill rakers count relationship is as follows in the sample examined: y = 1.4006x + 12.624. Coefficient of correlation, R amounts to 0.86 and is statistically significant, and the coefficient of determination, R² is amounted to 0.80. The the total length-gill rakers count relationship is presented in Fig. 2.

Fig. 2. Relationship between the gill rakers count and total length TL of the north African catfish

DISCUSSION

The catfish genus Clarias has a widespread distribution and is found in Africa and south-east Asia [28]. The most important economically species, C. gariepinus has been introduced in Europe, America and south-east Asia for aquaculture purposes [1].

The hybrids of C. gariepinus with other catfish species are also important economically. For example, in 1997 the north African catfish was introduced to Thailand and was successfully hybridized with Clarias macrocephalus for aquaculture [20]. Similarly, because the magur Clarias batrachus is less resistant to the wide fluctuations in temperature and oxygen, as well as it has lower growth rate and fecundity when compared to C. gariepinus, the successful works on hybridization among C. gariepinus and C. batrachus were conducted in Bangladesh [13]. Thus the proper identification within the Clarias genus may be complicated by both the individual variability and the human interventions, like manipulations in the aquaculture or hybridization [17].

The catfishes from the family Clariidae however are not only important economically, but also regarded as invasive species. As regards the genus Clarias, after Holčik [11] and according to the Convention of European Wildlife and Natural Habitats [8] two clariid species are present in Europe: C. gariepinus and C. batrachus (introduced mainly in the United Kingdom). Moreover, Gavriloaie and Chişamera [8] recorded the presence of a single specimen of the blunt toothed African catfish Clarias ngamensis in the lake in Bucharest  (Romania). Due to high predatory nature together with omni voracity and prolificacy C. gariepinus tends to impose a great thread to native fish fauna in Asia or Africa. Concern has also been raised against the possible introgression with Clarias species native in Asia, like C. batrachus and C. macrocephalus [18]. The introduction of C. gariepinus (as escapees from pounds built for recreational angling) was also recorded in the Guaraguaçu River basin in Parana State, Brazil, an area with an extremely rich endemic fish fauna, including many catfishes. C. gariepinus, very large and hard predators, posing serious potential for impact on the native fish fauna, needs study with emphasis on finding means for controlling its spread [33]. On the second hand, however, population control of species at lower trophic levels (such a tilapia Oreochromis niloticus) by culture with predators, such as the north African (= Nile) catfish C. gariepinus has been practiced worldwide. The proper use of predatory fishes considered as a safe biological method for covering tilapia overpopulation in ponds without affecting the big size prey [26]. Swingle [24] recommended the use of local predatory species for this purpose. The north African catfish is one of the most abundant and widely distributed fish in the river Nile and its tributaries, considered as the third important commercial fish in Egypt after tilapias and bagrids, cultivated under various systems [26].

According to Nei [20] native Thai species of Clarias catfishes – C. batrachus and C. gariepinus introduced to Thailand from Africa, can be easily distinguished by horizontal starch gel electrophoresis of allozymes. Genetic distances between species showed that C. batrachus and C. macrocephalus were the most similar species. The UPGMA dendrogram separated the samples into two main groups: one containing C. batrachus and C. macrocephalus, the other containing C. meladerma, C. gariepinus and Clarias sp. in the study of Na-Nakorn et al. [19]. The suggested relationship among C. batrachus, C. macrocephalus, and C. gariepinus  revealed by the UPGMA dendrogram agreed with the relationship based on meristic and morphometric data collected by Garcia-Franco [7].

The literature on the morphometric characters of the north African catfish is very scarce, including as a rule characteristics only of few characters. According to van Oijen [32] the north African catfish's maxillary barbels are rarely shorter than head, usually somewhat longer and usually longer in smaller specimens what agrees with the results obtained in the present paper (range – 93.8–142.3%, mean – 111.8%) of the head length. In the smallest length class the average was 119.8, while in the biggest length class in the sample studied – 102.3%. According to Kottelat [14] the dorsal head length (from tip of snout to extremity of occipital process) ranges within 26.3–34.5% of SL, what quite precisely agrees with the results of our study (28.2–35.8%, average 31.0%).

One of the most important morphological difference between C. gariepinus and other species within the genus Clarias, is the shape of the occipital process. For example, in C. batrachus the shape of occipital process is more or less triangular, its length is about 200% of its width [25]. The length of the occipital process in C. gariepinus constitutes 72.0–80.0% of its width in dependence of the fish length, decreasing slightly with the fish length, what was estimated in the own study.

In the paper of Turan et al. [30] the study of C. gariepinus from the river systems of Turkey showed the significant linear correlations between all morphometric characters and the fish length. The similar results were obtained in the present study.

Table 9 shows a comparative data on the morphology of three samples (in % of standard and head lengths) of the north African catfish, from Senegal [1], Turkey [3] and own study. The specimens examined from these samples were in the close range of the standard length SL, however fish from natural reservoirs were older (from I-IV years groups) than individuals from the heated water aquaculture in Poland (I year group), what indicates for much bigger rate of fish length under the condition of fishculture. The temperature of culture of the north African catfish in the Gołysz Institute of Ichthyobiology and Aquaculture fits within the temperature range in the natural habitat of the species in Senegal (up to 35°C) [1].

Table 9. Comparative data on morphology of three north African catfish samples (in % of SL and HL), from Turkey, Senegal and the sample investigated in this study

Symbol of character	Turkey – Göksu Delta [3] n = 49			Senegal [1] n = 49			Present study – heated water aquaculture, n = 100
	Range	Mean	SD	Range	Mean	CV	Range	Mean	CV
TL (mm)	135–324	–	–	–	–	–	90–260	–	–
SL – (mm)	110–283	–	–	133-183	–	–	84–222	156	25.4
% SL
HL	20.80–32.50	28.90	2.04	30.8-32.9	31.6	0.6	28.2-35.8	31.0	3.3
PPEL	15.30–31.54	21.46	2.23	22.7–25.9	23.9	0.8	22.1–26.6	24.4	3.8
PDL	32.28–47.54	35.92	2.43	32.2–38.1	35.7	1.5	25.3–38.3	36.1	4.0
PPL	21.41–54.16	45.75	4.86	46.1–51.4	48.8	1.8	43.8–54.2	47.5	3.0
PAL	51.4–70.67	56.64	2.97	56.0–59.9	57.9	1.3	51.9–62.6	57.1	2.7
DFL	40.74–67.80	62.20	3.89	59.3-65.1	62.6	1.9	58.7–65.1	61.9	2.2
AFL	37.14–45.61	42.37	1.84	39.3–46.0	43.0	2.3	38.4–46.3	42.9	3.4
% HL
IOW	38.90-59.20	43.50	3.14	33.0–39.8	37.5	1.9	32.4–41.2	37.2	4.5
OPW	–	–	–	18.3–22.4	20.9	1.2	14.7–24.0	19.7	7.8
OPL	–	–	–	14.6–19.4	17.5	1.4	11.8–19.4	14.7	10.0
PMW	–	–	–	20.0–23.9	22.3	1.0	23.5–32.7	28.4	5.9
VMV	–	–	–	20.3–22.7	21.5	0.7	17.6–26.9	23.4	6.3

A comparison of the morphometric characters described in present study and the data obtained by Agnese et al. [1] showed the results to be very similar, except for the length and width of the occipital process, being bigger in the north African catfish from the Senegal, and, on the opposite, the premaxillary toothplate and vomerine toothplate width being bigger in the sample in this study. The morphometric characteristics of the sample from Turkey [3] differed both from the sample from Senegal and own study (Table 9). The catfishes from Turkish waters have a smaller head, with the bigger interorbital distance and the pelvic and pectoral fins were located closer to the fish mouth.

Principal component analysis (PCA) performed by Turan et al. [30] on the catfishes from different regions of Turkey revealed that morphometric differentiation between samples from the Turkish waters was largely located in the head of C. gariepinus (length of head, interorbital distance, the distance from the dorsal fin to the occipital process), indicating this region to be important in the description of population characteristics, as well as the placement of dorsal and ventral fins. Results obtained in this paper showed also a high variability for the characters pertinent to head, but a low variability when regarded the placement of all fins.

Turan et al. [30] concluded, that such differences among the populations maybe related to different habitat characteristics, such as temperature, turbidity, food availability, water depth and flow, though it is a widely tolerant fish to extreme environmental conditions. It is well known that morphometric characters can show high plasticity in response to differences in environmental conditions, such as food abundance and temperature [30]. Extreme differences in culture conditions may elucidate adaptive potential of a species and its morphological variability as an expression of ecological plasticity [12].

Table 10 shows a comparative data on the meristics of three samples of the north African catfish, from West Africa [27], Turkey [3] and own study. Results obtained in this study in meristic characters showed that ranges of their values (except for the gill count) were wider, when compared to those reported by Teugels [27]. However the analysis of MFC ranges of particular meristic characters showed the very similar values. Thus the differences observed may be a result of higher number of the individuals examined in this study. In the case of the sample from Turkey [3], the fish studied showed a lower mean of the dorsal fin rays count and a considerably lower mean of the anal fin rays count, when compared to the results obtained in this study.

Table 10. Comparative data on meristics of three north African catfish samples from different regions of study

Symbol of character	Turkey – Göksu Delta [3]			West Africa [27]	Present study – heated water aquaculture
	Range	Mean	SD	Range	Range	Mean	SD
Db	50–83	64.66	8.55	61–68	55–75	67.42	3.5
Pb	7–9	9.00	–	–	6–11	8.86	0.7
Ab	30–53	43.41	7.07	45–65	36–59	51.29	3.7
GR	–	–	–	24–110	24–55	37.92	7.4
Vt.	–	–	–	56–63	52–64	58.37	2.2

Teugels [27] stated that gill rakers of the north African catfish are fine, long and closely set and their number increases with the size of fish. Gill rakers count is regarded by Teugels [29] as the most important and often the only reliable meristic feature to distinguish the species within the Clarias genus. The gill rakers ranges in this study and in the paper of Teugels [27] showed a clear overlap (Table 10).

CONCLUSIONS

The coefficient of correlation GR-TL, established in this study is very high and it is in evident agreement with the thesis by Teugels [27]. The coefficient of determination established in the present study showed that the total length determined the gill count as much as in 80%.

Meristic differentiation, slightly bigger in the case of the sample from Turkey, may resulted from the different number of individuals examined (much more numerous in the own study) and the different temperature of environment affected the development of meristic characters. The samples compared can be also differentiated genetically, bearing in mind that genetic variability of C. gariepinus is relatively high in comparison to other species [6,9]. The material of the north African catfish used in the European aquaculture is the most likely to origin from the African continent [29].

The morphological differentiation found in the dimensions of the premaxillary and vomerine toothplates between the African sample and the sample from own study, may result from the different type of fish feeding. The differences in the mouth structure between the domesticated and wild populations are well known in the literature [31].

REFERENCES

1. Agnese J.F., Teugels G.G., Galbusera P., Guyomard R., Volckaert F., 1997. Morphometric and genetic characterization of sympatric populations of Clarias gariepinus and Clarias anguillaris from Senegal. J. Fish Biol., 50, 1143–1157.

2. Chybowski Ł., 2000. Sum afrykański Clarias gariepinus. W: Ryby słodkowodne Polski. [The sharptooth catfish Clarias gariepinus. In: The freshwater fishes of Poland. 2^nd edition.] (ed.) Brylińska M. PWN, Warszawa, 360–361 [in Polish].

3. Ergene S., Portakal E, Karahan A., 1998. Karyological analysis and body proportion of catfish (Clariidae, Clarias lazera Valenciennes, 1840) in the Göksu Delta, Turkey.   J.  Zool., 23, 423–426.

4. FAO., 2000. FAO year book of Fisheries statistics. Aquaculture production. Vol. 90 /2. Food and agriculture, Organization of United Nations, Rome.

5. Filipiak J., Trzebiatowski R., Sadowski J., 1993. Effect of different FAT level in feeds upon some culture indices of African catfish (Clarias gariepinus) reared in cages in cooling water. Arch. Pol. Fish., 1(2), 113–123.

6. Galbusera P., 1997. The genetic variability of wild and inbred populations of the African catfish Clarias gariepinus (Burchell, 1822). Doctoral thesis, Katholieke Universiteit, Leuven.

7. Garcia-Franco M., 1993. Intra- and interspecific relationships of the Clariid catfish Clarias batrachus. PhD Thesis, Tokyo University of Fisheries.

8. Gavriloaie C., Chişamera Ch., 2005. Note on the presence of the blunt toothed African catfish Clarias ngamensis Castelnau 1861 (Pisces, Clariidae) in Romania. Travaux du Museum Natinal d'Histoire Naturelle "Grigore Antipa", Vol. 48, 309–315.

9. Giddelo C.S., Arndt A.D., Volckaert F.A.M., 2002. Impact of rifting and hydrography on the genetic structure of Clarias gariepinus in eastern Africa. J. Fish. Biol., 60(5), 1252–1266.

10. Holčik J., (ed.) 1989. The freshwater fishes of Europe. General introduction to fishes. Aula-Verlag Wiesbaden. V 1/II. 469.

11. Holčik J., 1991. Fish introductions in Europe with particular reference to its Central and Eastern part. Can. J. Fish. Aquat. Sci., 48, 13–23.

12. Keszka S., Krzykawski S., Więcaszek B., 2009. Variability of biometric characters of Acipenser baerii Brandt, 1869 in the heated water aquaculture. Electron. J. Pol. Agric. Univ. Ser. Fish  12(4).

13. Khan R., Mollah A., Ahmed U., 2000. Mass production of hybrid magur and its culture potential in Bangladesh. Aquaculture Res., 31(5), 467–472.

14. Kottelat M., 2001. Fishes of Laos. WHT Publications Ltd., Colombo 5, Sri Lanka. 198.

15. Lal K.K., Singh R.K., Mohindra V., Singh B., Ponniah A.G., 2003. Genetic make-up of exotic catfish Clarias gariepinus in India. Asian Fish. Sci., 16, 229–234.

16. Legendre M., Teugels G.G, Cauty C., Jalabert B., 1992. A comparative study on morphology, growth rate and reproduction of Clarias gariepinus (Burchell, 1822), Heterobranchus longifilis Valenciennes, 1840, and their reciprocal hybrids (Pisces, Clariidae). J. Fish. Biol., 40, 59–79.

17. Loy A., Molteni S., Bronzi P., 1997. Geometric morphometry in the characterisation of the cranial growth pattern of Acipenser naccari. In: Bronzi P., McKenzie D.J., DeCati G., Rossi R. (eds.). Proceedings of the 3^rd International Symposium on Sturgeon, 8–11 July, 1997, Piacenza, Italy. Blackwell Wissenschafts-Verlag. Berlin. Germany.

18. Na-Nakorn U., Hara M., Taniguchi M., Seki S., 1998. Isozymes variation of Clarias macrocephalus from four locations in Thailand. Fish. Sci. 64(4), 526–530.

19. Na-Nakorn U., Sodsuk P., Wongrat P., Janekitkarn S., Bartley D.M., 2002. Isozyme variation among four species of the catfish genus Clarias. J. Fish. Biol., 60(4), 1051–1057.

20. Nei M., 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 23, 341–369.

21. Nowak M., Szczerbik P., Tatoj K., Popek W., 2008. Non-native freshwater fishes in Poland: an overview. AACL Bioflux. 12, 1(2), 173–191.

22. Nukwan S., Tangtrongpiros M., Lawanyawut K., Veerasidth P., 1990. Hybridization between Clarias macrocephalus and Clarias gariepinus. Proceedings of the 28^th Kasetsart University Conference. Bangkok: Kasetsart University Press (in Thai), 553–568.

23. Ruszczyc Z., 1981. Metodyka doświadczeń zootechnicznych [Methodology of husbandry experiments]. PWRiL, Warszawa [in Polish].

24. Swingle H.S., 1960. Comparative evaluation of two tilapias as pond fishes in Alabama. Transac. American Fish. Soc., 8, 142–148.

25. Taki Y., 1974. Fishes of the Lao Mekong basin. United States Agency for International Development Mission to Laos Agriculture Division. 232.

26. Tawwab M. A., 2005. Predation efficiency of Nile catfish, Clarias gariepinus (Burchell, 1922) on fry Nile tilapia Oreochromis niloticus (Linnaeus, 1758): effect of prey density, predator size, feed supplementation and submerged vegetation. Turkish J. Fish. Aquatic Sci., 5,  69–74.

27. Teugels G.G., 1986. A systematic revision of the African species of the genus Clarias (Pisces, Clariidae). Ann. Mus. R. Afr. Centr. Sci. Zool., 247, 1–199.

28. Teugels G.G., 1996. Taxonomy, phylogeny and biogeography of catfishes (Ostariophysi, Siluroidei): an overview. Aquat. Living Resour. 9, 9–34.

29. Teugels G.G., 1998. Intra- and interspecific morphometric variation in Clarias gariepinus and C. anguillaris. (Siluroidei, Clariidae). Genetics and Aquaculture in Africa. (Ed.),  J.F. Agnese, ORSTOM (Paris), Collection Colloques et Séminaires, 241–247.

30. Turan C., Yalçin Ş., TuranF., Okur E., Akyurt I., 2005. Morphometric comparisons of African catfish, Clarias gariepinus, populations in Turkey. Folia Zool. 54(1–2), 165–172.

31. Wimberger P.H., 1992. Plasticity of fish body shape – the effects of diet, development, family and age in two species of Geophagus (Pisces: Cichlidae). Biol. J. Linn. Soc., 45, 197–218.

32. van Oijen M.I.P., 1995. Appendix I. Key to Lake Victoria fishes other than haplochromine cichlids. In: F. Witte and W.L.T. van Densen (eds.). Fish stocks and fisheries of Lake Victoria. A handbook for field observations. Samara Publishing Limited, Dyfed, Great Britain, 209–300.

33. Vitule J., Umbria S., Aranha J., 2006. Introduction of the African Catfish Clarias gariepinus (BURCHELL, 1922) into southern Brazil. Biol. Invasions. 8(4), 677–681.

 

Accepted for print: 30.03.2010

---

Beata Więcaszek
Department of Fish Systematics,
West Pomeranian University of Technology, Szczecin, Poland
Kazimierza Królewicza 4, 71-550 Szczecin, Poland
email: bwiecaszek@zut.edu.pl

Stanisław Krzykawski
Department of Fish Systematics,
West Pomeranian University of Technology, Szczecin, Poland
Kazimierza Królewicza 4, 71-550 Szczecin, Poland

Artur Antoszek
Department of Fish Systematics,
West Pomeranian University of Technology, Szczecin, Poland
Kazimierza Królewicza 4, 71-550 Szczecin, Poland

Joanna Kosik
Department of Fish Systematics,
West Pomeranian University of Technology, Szczecin, Poland
Kazimierza Królewicza 4, 71-550 Szczecin, Poland

Paweł Serwotka
Department of Fish Systematics,
West Pomeranian University of Technology, Szczecin, Poland
Kazimierza Królewicza 4, 71-550 Szczecin, Poland

---

Responses to this article, comments are invited and should be submitted within three months of the publication of the article. If accepted for publication, they will be published in the chapter headed 'Discussions' and hyperlinked to the article.

---

Main  -  Issues  -  How to Submit  -  From the Publisher  -  Search  -  Subscription