Katle (Neolissocheilus hexagonolepis (McClelland) reproduction IN the Indrasarobar reservoir and the Tadi River, Nepal. (by D.B. Swar and J.F. Craig)

D.B. Swar
Directorate of Fisheries Development, Central Fisheries Building Balaju, Kathmandu, Nepal
and
J.F. Craig
Whiteside, Dunscore, Dumfries, Scotland, DG2 OUU

ABSTRACT

A comparative study on the reproductive biology of katle, Neolissocheilus hexagonolepis (McClelland), was carried out in the Tadi River and in the newly created Indrasarobar reservoir, Nepal. The intensive field survey and laboratory examination revealed that the gonads started to develop in March and the reproductive phase lasted from April to October, which is consistent with the reproductive cycle of the other cyprinids. Katle from the river were less fecund than the reservoir katle at smaller size, however, absolute fecundity seemed to increase with size more rapidly in the river fish, and they were more fecund than the reservoir fish at the higher total length. The average relative fecundity in the reservoir population was 19.13 ±1.56 eggs g-1 compared to 22. 57±1.41 eggs g-1 for the riverine population. Occurrence of different size groups (batches) of oocytes indicates the multiple spawning characteristic of katle. The upstream migration of ripe females in the reservoir and river suggested that katle was very selective in finding a suitable place to spawn.

1. INTRODUCTION

Studies on the reproduction of cyprinids on the Indian subcontinent have concentrated mainly on two groups, the Indian major carps, Labeo spp, Cirrhinus spp and Catla catla, and the hill stream fish Tor spp (mahseer), Neolissocheilus hexagonlepis (katle) and Schizothorax spp (snow trout). The Indian major carps tend to be single spawners while the latter are mainly multiple spawners.

Despite interest in economically important cyprinids and the development of their aquaculture, the reproductive biology of katle is poorly known. Langdale Smith (1944) carried out one of the earliest documented observations on the reproductive ecology of katle. Ahmed (1948) collected and incubated the fertilised eggs of katle and studied their embryonic development. Rai (1978) stripped eggs from mature katle captured in the Trishuli River, central Nepal. The eggs were fertilised and incubated at 18-21°C. Dasgupta (1988) estimated the fecundity of katle collected from the Simsang River, India. He found that the katle was a prolific breeder but less fecund than mahseer.

The present study investigated the reproductive adaptability of katle by comparing its reproductive season, spawning behavior and migrations, and fecundity in the Tadi River with those in the Indrasarobar reservoir, Nepal.

2. MATERIAL AND METHODS

Katle were collected from the Indrasarobar reservoir and the Tadi River from January 1988 to December 1990 (Swar, 1994). The fish were sexed from external and internal examination. Total length (mm) and weight (g) were recorded. Pectoral fin rays were used to age the fish (Swar, 1994). The gonads were removed and weighed and the stages of development were noted (Laevastu, 1965). The outer ovarian walls were removed and the ovaries stored in Gilson's fluid (Bagenal, 1971) for about one month. After cleaning, three egg fractions, according to diameter, were collected using sieves with mesh sizes of 1 000 and 1 700 ìm. From each fraction, four sub-samples of 1000 eggs were air-dried for 48 h together with the remaining eggs. Total numbers of eggs were calculated from the weight of the known number and the total weight. All eggs were counted when the total was <4000. Fecundity was determined for 84 ripe females from the Indrasarobar reservoir and 35 from the Tadi River.

To determine the size of ova at different stages of maturity, 200 preserved eggs were randomly selected from both the anterior and posterior parts of the ovary in the second, third, and fourth stages of maturity from both the reservoir and river fish. Egg diameters were measured using an ocular micrometer. To examine the size range of ova in mature ovaries, 100 eggs from each of the three fractions were measured.

Radio tags (49 MHZ) were used to investigate the spawning migration and to locate the breeding sites of reservoir female katle. Four sexually mature, spawning females were collected from the reservoir during 26-27 August 1990. Fish were 360-520 mm in total length and 450-1400 g in weight. They were anaesthetised with benzocaine (25 mg/L) (Laird and Oswald, 1975), transmitters were placed beside the dorsal fin (Winters, 1978) and then all were released at the same place. The fish were followed with a radio receiver either from land or from a boat between 26 August and 15 September 1990.

The spawning behaviour of katle was observed in the Tadi River in 1991. The Khahare Khola, an afferent stream of the Tadi River, had already been identified as a spawning site (personal communication with local fishers). Intensive observations were conducted from 15 August to 7 September 1991. In addition fyke nets were set in the Khahare Khola near its confluence with the Tadi River. The nets were checked at intervals of 4 hours from 1900 to 0600 h. Water temperature, pH, turbidity and dissolved oxygen were measured during the period of observation. The shallow areas with gravel bottom were surveyed for fish eggs once a day using a dip net.

3. RESULTS

3.1 Sex ratios

The proportion of females was significantly higher than that of males (p<0.05) in the reservoir samples in all years studied. Males were more numerous in the river sample in 1988 (p<0.05), but there were no differences in the sex ratios in 1989 and 1990 (Table 1). The sampling areas in both habitats were extended during the second half of 1989 and in 1990. The upper area of the Indrasarobar reservoir was fished with different gears such as cast net and local traps and the area covered in the Tadi River was extended by 2 km to the north and south beyond the normal 10 km stretch. Although the abundance of the males in the reservoir increased slightly and the proportion of females increased in catches from the river, the ratios did not change significantly.

3.2 Maturity stages

Seven stages of gonad development were identified for female and male katle from both habitats (Table 2). At the first and second stages of maturation, the ovaries appeared as two small organs of compact tissue under the air bladder, and the oocytes were not visible to the naked eye. The oocytes became visible at stage three. The ovaries gradually increased in size and the oocytes became more distinct and translucent at successive stages. At stage 1, the testes were very thin, thread like, paired organs under the vertebral column. They gradually increased in size and ultimately occupied about two thirds of the ventral cavity. The margin became convoluted with transverse grooves. At stages 5 and 6, the testes were pinkish white and turgid, and milt ran with slight pressure. The 'partly spent' stage expected to be found in a multiple spawner like katle was not found in either sex.

Table 1

The sex ratios of katle in the reservoir and river

Table 2

Stages in the development of the ovaries and testes of katle

Stage 1: State: virgin:

Ovaries: Thin, small, two pieces of compact tissue underneath the air bladder, gray. Oocytes not visible to the naked eye.

Testes: Very small paired organs, close under the vertebral column, connected to the air bladder, transparent.

Stage 2: State: Maturing virgin and recovering spent:

Ovaries: Ovaries with compact lobes, creamy to pale yellow, ova spherical, particularly laden with yolk, eggs visible with magnifying glass.

Testes: Slightly larger than stage 1, length about _ of the ventral cavity, transparent, reddish gray.

Stage 3: Developing:

Ovaries: Light yellow with distinct blood capillaries, eggs opaque, visible with naked eye, whitish and granular.

Testes: Opaque and reddish with blood capillaries, occupy about half the length of the ventral cavity.

Stage 4: Developed:

Ovaries: Large and bright yellow with conspicuous blood capillaries and oval shaped oocytes.

Testes: Reddish white with wavy margins and distinct blood capillaries, occupy about 2/3 of the ventral cavity.

Stage 5: Gravid:

Ovaries: Enlarged and fill ventral cavity, yellow eggs completely round and semi-transparent; primary oocytes as in stages 1 and 2 present.

Testes: White, waxy margin with transverse grooves; occupy the length of the ventral cavity; drops of milt exuded on pressure.

Stage 6: Spawning:

Ovaries: Distended, yellowish white, jelly-like, eggs run with slight pressure, most eggs translucent.

Testes: Occupy the length of the ventral cavity; pinkish white, turgid and milt runs with slight pressure.

Stage 7: Spent:

Ovaries: Empty, shrunken, baglike, a few residual oocytes may be visible.

Testes: Flaccid, gray, without milt.

3.3 Maturation cycle

Since the annual maturation cycle of the katle appeared to be regular, the years 1988 to 1990 for each sex and each habitat were pooled. Reservoir females at stages 1 and 2 were found throughout the year. More than 60% of the katle were at stage 2 except in March. In March stages 3 and 4 were observed and all stages, 1 to 6, were present from April to October. Gravid females were caught from April to October but formed a small percentage of the total catch. No running or spent females were ever found.

As in the females, stage 2 development was dominant among the reservoir males and all fish were either at stages 1 or 2 from November to February. Stage 3 was observed every month from March to October except May. Males at stage 4 were present in the reservoir in April, July and August. Males of stages 5 were found only in July and August. No males of stages 6 and 7 were ever caught in the reservoir.

The general cycle of gonadal maturation of the katle in the Tadi River was similar to that in the Indrasarobar reservoir. Females and males at stage 2 were found throughout the year. Females at stage 3 appeared in March and were frequently caught until October. Stage 5 was observed from July to October and stage 6 from September to October. Males at stages 3, 4 and 5 were found from March until October. Spent males and females were encountered only in September and October, respectively.

3.4 Age at first maturity

About 50% of the river males were mature at age 2 yr and >90% at 3 yr; only 5% of river females were mature at 2 yr but nearly 100% were adult at 3 yr. No male fish <3 yr were caught in the reservoir. Most of the 3-yr-old males there were mature. In the reservoir about 50% of 3-yr-old and 100% 4-yr-old females were mature. The immature fish were significantly smaller than mature fish of the same age in both sexes in both the reservoir and river populations.

3.5 Fish movements

Four mature female fish fitted with radio tags, released at point A (Map 1) on 26 and 27 August 1990, remained at first near the release site (Table 3). All then migrated upstream. Three moved towards the Kulekhani River and the fourth towards Chitlang Khola.

Fig. 1 - Map of Indrasarobar reservoir showing the areas where radio tagged fish were located

Table 3

Movements of katle as determined by radio tagging in the Indrasarobar reservoir. R=release (date and place). For locations see Map 1 (* = not located)

3.6 Visual observations of katle spawning behavior

The upstream movement of 3 fish to the Khahare Khola stream from the Tadi River was observed on 20 August at about 1600 h. They appeared to move rapidly for the first 300 m. They slowed down after they reached the shallow and slow moving part of the stream. The largest fish, probably female, was leading and two smaller fish, presumably males, followed her. As soon as they found a wide, shallow pool with a gravel bottom, they stopped. They then started to swim in a non-directional manner, over and around the spawning substratum. Some movements were circular (anti-clockwise and clockwise), whereas others were spiral. Both the males were seen to be equally active and swam alongside the female. After _ h the same fish were noticed entering the Khahare Khola. Two spent fish, one of each sex were later caught in a fyke net. Eggs were found on the gravel bottom of the spawning site the next morning. One thousand fertilised ova were collected and incubated at the Fishery Research Station, Trishuli. The water temperature at the spawning ground ranged from 20-22°C, the mean dissolved oxygen was 6.98 ± 0.98 mg 1-1, and pH 6.73 ± 0.11.

3.7 Egg diameter and fecundity

At stage 2, one batch of oocytes was present, at stage 3, there were two distinct batches and at stage 4, three types, immature, maturing and mature. Data on the diameters of mature eggs at >1.60 mm, and on the diameters of mature eggs at stage 5 from the reservoir and river from July to October were analyzed by ANOVA. The location on its own did not appear to influence the size of the eggs, but the time of maturation (month) was statistically significant. The largest eggs were produced in September. Absolute fecundity of katle in both the river and reservoir varied considerably amongst individuals. It ranged from 1 387 (in a fish of length 160 mm and weight 53.8 g) to 33 270 eggs (535 mm and 1750 g) in the reservoir population, and from 760 (155 mm and 40 g) to 8 951 (395 mm and 500 g) in the river population. The mean relative fecundity in the reservoir population was significantly different (p < 0.01) from that of the river, 19.13±1.56 compared to 22.56±1.41 eggs per gram, respectively.

4. DISCUSSION

Males were normally more numerous in the river catches. This trend has been reported in other stream-living cyprinids (Siddique et al., 1976; Mann, 1980; Cooper, 1983). The imbalance was probably related to differences in age-related mortality and spawning migrations. Al-Kholy (1972) observed that Puntius barberinus females stayed longer in the spawning areas than males and this significantly reduced the proportion of females in the main population. In this study, katle were found to have a prolonged breeding season from April to October and most of the mature females probably migrated to suitable breeding grounds and remained there longer than the males. It seems that the higher ratio of males to females during 1988 was the result of the upstream spawning migration of the females in the Tadi River, beyond the sampling area. This was confirmed by the increased proportion of females during the second half of 1989 and in 1990 when the sampling areas were extended by 2 km. The extremely low number of male fish in the reservoir is difficult to explain. Males may prefer to live in a riverine environment and reside in the Kulekhani River. The males appear to have a higher mortality rate so there were fewer older males. No males >350 mm were caught in the reservoir. The use of different sampling gears may have been a reason for the variability in the proportion of males in the two habitats. The reservoir fish were sampled using multi-mesh gill nets and the smallest fish caught were 130 mm. The smallest males caught in the river were 80 mm. Therefore, it seems possible that males between the sizes of 80 and 130 mm constituted a higher percentage of the reservoir katle.

Temperature and photoperiod modulate sexual development and spawning in cyprinids, although the former is the predominant factor in their reproductive cycle between November and February. The gonads started to develop in March and the main reproductive phase lasted from April to October. The mean surface water temperature in both the reservoir and river began to rise in March, decreasing again in November and reaching its lowest levels in January and February (Swar, 1994). Mean day-length varied from 625 minutes in December to 831 minutes in June. From March to October longer day-length and higher temperatures coincided with increased reproductive activities. This is consistent with the reproductive cycles of other cyprinids (Hyder, 1970; Abidin, 1986). Ahmed (1948) observed that katle bred from April to October with a peak in August and September.

As katle has a prolonged spawning season, which extends before and after the monsoon (May to September), flooding does not seem to be an essential factor. However, katle may take advantage of flood water to reach breeding areas. Desai (1973) recorded migrations of Tor tor during monsoon floods in the River Narmada, India. The spawning movements of ripe females in the reservoir and river suggested that katle was very selective in finding a suitable place to spawn.

The percentage of different size groups (batches) of oocytes indicates the multiple spawning characteristics of katle. In several studies on the fecundity of multiple spawners, attempts have been made to estimate both batch size and number of batches (Mackay and Mann, 1969; Macer, 1974). Batch size can be defined as the number of oocytes released together. The number of batches during the spawning period depends on temperature, photoperiod and food supply, which will affect the physiological state of the fish. In multiple spawners, egg release would appear to be very flexible, since the process of synchronous oocyte development and oocyte reabsorption makes it possible to control egg numbers. Nikolsky (1969) stated that multiple spawning was only successful when there was a long period of adequate food supply for the larvae. However, he warned that it would be wrong to regard this life history strategy as an adaptation to food supply alone because it could also be an adaptation to unsuitable spawning conditions. Burt et al. (1988) suggested that multiple spawning within a season was associated with less seasonal variation (single spawners were constrained by more severe environmental conditions), small body size and small relative ovary size. Although these conditions may be true for fish in general, they do not seem to be applicable to katle in central Nepal which are subjected to seasonal changes in environmental conditions and whose body size is relatively large. In this low fecundity species, multiple spawning may have developed as an adaptation (Swar, 1994). Extreme conditions such as flooding in the fast flowing rivers caused by monsoon rains and the substantial annual drawdown in the reservoir probably cause these mortalities.

Acknowledgements

The authors wish to thank Mrs. H.J. Craig for her useful suggestions during the preparation of this manuscript. The study was carried out during the tenure of an International Development Research Centre fellowship to the first author.

References

Abidin, A.Z., 1986. The reproductive biology of a tropical cyprinid, Hampala macrolepidota (van Hasselt), from zoo Negra lake, Kuala Lumpur, Malaysia. J. Fish Biol. 29: 381-391.

Ahmed, N., 1948. On the spawning habits and early development of copper mahseer Barbus (Lissocheilus) hexagonolepis (McClelland). Proceedings of the National Institute of Science of India 14: 21-28.

Al-Kholy, A.A, 1972. Aquatic resources of Arab countries. Alesco Science Monograph series. 451p.

Bagenal, T.B., 1971. Eggs and early life history, part I. Fecundity. In: Ricker, W.E. (ed.): Methods for the Assessment of Fish Production in Fresh Waters: 112-131. I.B.P. Handbook No. 3. Blackwell Scientific Publications, Oxford.

Burt, A., K.L. Kramer, K. Nakatsuru and C. Spry, 1988. The tempo of reproduction in Hyphessobrycon pulchripinnis (Characidae), with a discussion on biology of multiple spawning in fishes. Env. Biol. of Fishes 22: 15-27.

Cooper, A., 1983. The reproductive biology of poor-cod, Trisopterus minutus L., whiting, Merlanguis merlanguis L., and Norway pout, Trisopterus esmarkii Nilsson, off the west coast of Scotland. J. Fish Biol. 22: 317-355.

Dasgupta, S.M., 1988. A study on the fecundity of Acrossocheilus hexagonolepis (McClelland) from Garo hills, Meghalaya, Uttar Pradesh. J. Zool. 8: 159-167.

Desai, V.R., 1973. Studies on the fishery and biology of Tor tor (Hamilton) from the River Narbada. J. Inland Fisheries Soc. India 2: 101-111.

Hyder, M., 1970. Gonadal and reproductive pattern in Tilapia leucosticta (Cichlidae) in equatorial lake Naivasha (Kenya). J. Zool. Soc. London 162: 179-195.

Laevastu,T., 1965. Manual of methods in fishery biology. FAO Manuals in Fisheries, series number 1: 1-9.

Laird, L.M. and R.L. Oswald, 1975. A note on the use of benzocaine (ethyl p-aminobenzoate) as a fish anaesthetic. Fisheries Management 6: 92-94.

Langdale Smith, W.K., 1944. A note on the breeding habits of Katli, Barbus (Lissochilus) hexagonolepis McClelland. J. Bengal Natural Hist. Soc. 19: 89-90.

Macer, C.T., 1974. The reproductive biology of horse mackerel Trachurus trachurus (L.) in the North Sea and English Channel. J. Fish Biol. 6: 415-438.

Mackay, I. and K.H. Mann, 1969. Fecundity of two cyprinid fishes in the River Thames, Reading, England. J. Fisheries Res. Bd. Canada 26: 2795-2805.

Mann, R.H.K., 1980. The growth and reproductive strategy of the gudgeon, Gobio gobio (L.) in two hard-water rivers in southern England. J. Fish Biol. 17: 163-176.

Nikolsky, G.V., 1969. Fish Population Dynamics. Oliver & Boyd, Edinburgh.

Rai, A.K., 1978. Artificial breeding of katle, Acrossocheilus hexagonolepis. J. Nat. Hist. Museum of Kathmandu 2: 45-50.

Siddique, A.Q., A. Chatterjee and A.A. Khan, 1976. Reproductive biology of carp, Labeo bata (Ham.), from River Kali, India. Aquaculture 7: 181-191.

Swar, D., 1994. A study on the ecology of katle, Neolissocheilus hexagonolepis (McClelland) in a Nepalese reservoir and river. PhD thesis, The University of Manitoba, Winnipeg, Manitoba. 261p.

Winters, J.H., 1978. Healing of wounds and location of tags in Atlantic herring (Clupea harengus harengus) released with abdominally inserted magnetic tags. J. Fisheries Res. Bd. Canada 34: 2402-2404.