Age and growth in the Indian major carp Labeo rohita ( Cypriniformes : Cyprinidae ) from tropical rivers of Ganga basin , India

Edad y crecimiento de la carpa mayor de la India Labeo rohita (Cypriniformes: Cyprinidae) de los ríos tropicales de la cuenca Ganga, India. Labeo rohita is a member of the Indian major carp species and originally an inhabitant of the Ganga river network in India. It is among the top ten aquaculture species of the world. Since there is a lack of information on the growth pattern of the wild populations of this species, this study aimed at evaluating the pattern of age and growth, to support the development of effective management plans. A total of 1 082 samples of L. rohita were obtained from May 2009 to July 2012 in six drainages of the Ganga basin. Scales of the fish were used to determine the age and growth by analyzing the annual rings growth. Out of six populations, a maximum 8+ age classes were recorded from two rivers (Betwa and Sharda). The back-calculated lengths at 8+ age class ranged from 86.22cm to 91.66cm. However, for the rest of rivers up to 7+ age classes were recorded. Among growth parameters, specific rate of length increase (Cl) and specific rate of weight increase (Cw) showed decreasing trend, and three distinct life stages of L. rohita were recorded based on growth constant data (Clt). Analysis of variance (ANOvA) of the back-calculated length data set of six populations indicated a significant difference (p<0.05). All three populations showed significant variation in length attainment during 1+ and 3+ age groups, while two populations showed significant variation in length attainment during the 2+ and 7+ age classes. Additionally, analysis of age frequency at different length intervals indicated that with increase in age class, number of fish samples was reduced. Since the pattern of life history traits of L. rohita have not been attempted in the recent past; therefore, this study will guide fisheries biologists about the current stock structure of this fish across different spatial scale of the Ganga basin. Rev. Biol. Trop. 61 (4): 1955-1966. Epub 2013 December 01.

The analysis of life history traits has been widely used by ichthyologists to differentiate among different species and different populations within a species; fortunately, this method continue to be used successfully (Ihssen et al., 1981).Studies of morphologic variation among populations still have an important role to play in stock identification, despite the advent of biochemical and molecular genetic techniques, which accumulate neutral genetic differences between groups (Mir, Sarkar, Dwivedi, Gusain & Jena, 2013).Studies on age and growth, maturity of commercially important fishes, provide baseline information that typically assists with the initial recognition and delineation of geographic regions that are representative of individual stocks (Pawson & Jennings, 1996) and is an almost essential prerequisite for successful stock identification (Griffiths, 1997).The use of such parameters is an efficient and cost-effective tool for stock identification, as these data are routinely collected for assessment and management purposes (Ihssen et al., 1981).Although the utility of these parameters for stock identification appears to decrease with stock complexity, their applicability increases with the number and diversity of parameters examined (Begg & Waldman, 1999).
Among Indian fish species, age and growth aspects of the two major carps of India, Cirrhinus mrigala and Catla catla have been investigated previously (Khan & Siddiqui, 1973;Sarkar, Negi & Deepak, 2006;Mir et al., 2012).Nevertheless, there is a need of scientific information on age and growth profile of L. rohita, the most important carp in India.L. rohita (Hamilton, 1822) commonly known as 'Roho labeo' thrives in lakes, ponds and rivers in India and adjacent countries (Talwar & Jhingran, 1991).The natural resources of this fish extends from the network of the Ganga, the Sindh and the Brahmaputra river systems in the North, and the East and West coast river systems flowing through in the South and Central India.This species of fish grows up to 200cm size (Frimodt, 1995); however, a considerable decline in the overall size and species in natural waters (Mir et al., 2013) has been observed, and the species is now categorized as LC (Least Concern) as per IUCN (2012).
Recently, the National Bureau of Fish Genetic Resources (NBFGR), Lucknow, India, has started a flagship network programme on the stock identification of L. rohita (Lakra & Sarkar, 2010).Besides, to successfully develop and manage the population of L. rohita in natural waters of the Ganga basin, it is important to understand the current pattern of biological traits in natural habitats.Thus, this study was undertaken to provide information about the important life history attributes (age and growth) of L. rohita, and to compare these aspects among different rivers of the Ganga basin, that may support effective conservation and sustainable management actions for the stocks of L. rohita in the area.

Study area:
The river Ganga basin is located 22°30'0'' -31°30'0'' N -73°30'0'' -89°0'0'' E. Total drainage area exceeds 1 060 000km 2 , the basin is the fifth largest in the world.The length of main channel from the traditional source of the Gangotri Glacier in India is 2 550km.The annual volume of water discharged by the Ganga is the fifth highest in the world, with a mean discharge rate of 18.7 x 103m 3 /sec (Welcomme, 1985).The main sources of water in the basin are direct seasonal rainfall, mainly from the South-West and glacial and snowmelt during the summer (Chapman, 1995).The main channel of the Ganga begins at the confluence of Bhagirathi and Alaknanda, which descend from upper Himalayas to Devprayag (520m.a.s.l.) and receives a number of major tributaries.The Northern tributaries that enter on the left bank principally include Ghagra, Gomti, Buri Gandak and Kosi and Southern tributaries include Yamuna (Ken, Betwa), Son, Chambal, Tons, Kalisindh, Sharda and Damodar.Most of these tributaries are controlled by irrigation barrages and there are two major barrages across the main channel, one at Hardwar which abstracts much of the water at this point to irrigate the Doab area, and one at Farakka which diverts water down to Calcutta.All of these structures modify the flow of the river and may considerably influence fish distribution.Fish and fisheries are both important resources and activities in their own right, but also provide indicators of the overall impact of anthropogenic changes over the basin.
Sample collection: Samples of L. rohita were collected during May 2009 to July 2012 from six different rivers of the Ganga basin (Fig. 1).The GPS coordinates, implicating factors of the selected sites on these rivers, and the number of samples collected from each river is shown in Table 1.The samples were collected from the selected rivers (upstream, midstream and downstream) in the morning before the sunrise, by using cast nets (9m length, 9m breadth and 1/2cm mesh size) and drag nets (100m length, 20m breadth, 1⁄2cm mesh size), and also from the 18 landing centers present on these six rivers.Measurements on fish length and weight were taken; besides, fish were dissected and sexes identified per collection site.Total length of each fish was taken from the tip of snout (mouth closed) to the extended tip of the caudal fin, nearest 0.01mm by digital caliper and weighed to the nearest 0.01g (total weight) (by digital weighing machine, ACCU-LAB Sartorius Group).

Analysis of age and growth by scale preparation and reading techniques:
To study age and growth, four to five scales of each fish were taken, from the area below the posterior margin of the dorsal fin, and two scale rows above the lateral line from each fish, by using a scalpel.After collection, the scales were washed with 5% KOH and rinsed in distilled water.Scales were cleaned and mounted dry between two glass slides (4 x 4") for permanent storage and studied under profile projector (Sipcon made) at appropriate magnification.To avoid any biases, scale reading was performed by same person every time.The true mark or annuli (defined by light bands in the form of grooves and extending around the scale) were included in the age determination; whereas, false annuli (or marks formed due to unfavorable condition like unavailability of natural food, environmental alteration leading to cessation of growth), were excluded from the analysis.The total radius of the scale (S) and scale radius from nucleus to the annulus (S n ) were measured using K25d-img software connected to the profile projector.Growth patterns were determined using size at age data from counts of inner layer in scales.Since scales have been found very convenient and authentic in use for age and growth (Khan & Khan, 2009), the following formula given by Bagenal & Tesch (1978) have been used for back calculation based on the scale readings as: Where, a=correction factor L=Total length (mm) of the fish at the time of scale removal.
L n =Total length (mm) of the fish at the time of annulus "n" formation S n =Scale radius (mm) from nucleus to the annulus "n".S=Total radius (mm) of the scale.
Specific rate of linear growth (C l ), Specific rate of weight increase (C w ), Growth constant (C lt ) and specific average size (h) were derived as per Chugunova (1963).Specific rate of linear growth (C l ): Where L n =Mean computed total length of fish at ultimate years of life L n-1 =Mean computed total length of fish at penultimate years of life.Specific rate of weight increase (C w ): Where W n =Mean computed total weight of fish at ultimate years of life.W n -1=Mean computed total weight of fish at penultimate years of life.Growth constant (C lt ): C lt =log L n -log L n-1 /0.4343*t 2 +t 1 /2 t 2 and t 1 are time intervals between ultimate and penultimate age groups and the value of t 2 +t 1 /2 is equal to 1.5.Here age groups were defined as the beginning and end of a year group and denoted as 1, 2, 3 and so on.
An analysis of variance was carried out to detect significant difference of six different populations by using back-calculated length data of each year (L1 to L8).The statistical analysis was carried by using ORIGIN 6.1, SPSS 12.0 and EXCEL 2007.

RESULTS
A total of 1 082 fishes were collected.The scales of L. rohita were cycloid and were distinguished into true ring, false ring and larval marks besides radii.Annual growth rings were clearer and sharper in scales thereby producing lesser errors in age estimation (Fig. 2).The focus, which represents the initiation of growth in scales, is located near the anterior field of the scale; while, the larval marks appear in the first year age class and are situated not far from Betwa and lowest in river Sharda.The analysis of the back-calculated results indicated variation at different age classes among the studied populations (Fig. 3).
On the basis of above results, the annual rate of growth (h) over the years was calculated for different locations.Maximum average growth indicated higher values for the river Ken (173.67)followed by river Ghagra, Sharda, Gomti and Ganga, while minimum growth increment was observed in river Betwa (110.80).The growth decreased with increasing age except river Betwa, Ganga, Sharda and Gomti, where growth suddenly increased at 5+ age class of river Betwa and Sharda and 7+ age class of river Betwa, Ganga, Sharda and Gomti.In order to detect population groups of L. rohita with similar growth patterns, we employed Bray-Curtis distance and the unweighted pair-group method using arithmetic averages (UPGMA) in our application of a hierarchical cluster analysis to six rivers × six length at-age (one to six years) matrix.The six populations resolved into two groups (Fig. 4) (by decreasing order of growth rate):  group one included river Ken that supported the population with the highest growth rate in the study area; group two, comprised river Sharda, Ganga, Betwa, Gomti and Ghagra.An analysis of variance was used to detect significant differences between the locations.ANOvA was carried out by using observed back-calculated length data of each year (L1 to L8) of six different populations (Table 2).Significant interspecific variation was recorded in length among locations.Out of six populations, three showed significant variation in 1+ and 3+ age group whereas, only two showed significant variation within different locations in 2+ and 7+ age class.Analysis of backcalculated length of 1+ age class indicated significant variation for some of the population.River Sharda showed significant difference with river Ganga and river Ghagra.There was also a significant variation between river Sharda and river Ganga at 2+ age class.At 3+ age class, river Sharda showed significant difference with river Ganga and river Ghagra.Significant variation was observed between river Sharda and river Ganga at 7+ age class, non-significant variation was observed 4+, 5+, 6+ and 8+ age class.
Among growth parameters, specific rate of linear growth (C l ) and specific rate of weight increase (C w ) decreased for all populations.However, C l increased at 5+ and 7+ age class in river Betwa and Sharda whereas, C w increased at 7+age class in river Betwa.
Based on this analysis it was concluded that C l exhibited higher values for the population of river Sharda followed Betwa, Gomti, Ken, Ghagra and Ganga at 1+ age class.At 7+ age groups, it was maximum for river Sharda and minimum for river Betwa.Similar pattern of variation was also observed at rest of the age classes (Fig. 5A, B).The index of population weight intensity (ØC w ) of L. rohita was maximum for river Ken (170.50)followed by river Ghagra (160.25),river Sharda (133.43),river Betwa (99.71), river Ganga (97.67) and river Gomti (93.30).
The growth characteristics (C lth ) are useful to determine the period, where the first year ends and the second phase begins.Growth constants (C lt ) are useful to determine the periods of life span of the fish.The results of the present study clearly indicate that L. rohita enters second period of life after third year in all the studied rivers (Fig. 5C).The analysis of the growth constant indicated maximum three life phases from all the studied rivers except river Ken where two life phases were observed.Additionally, the percentage of age frequency was estimated at different total length intervals among six populations, and it was observed that with an increase in fish total length, the higher age classes declined, and with smaller length intervals, lower age classes were importantly present (Fig. 6).

DISCUSSION
The present study reported the age and growth composition from six different populations of L. rohita in Ganga basin, and showed considerable intra-basin variations.We have observed that the growth increment in terms of length attainment in the successive years varied in different rivers.Lower growth increment was recorded for the samples of river Betwa, whereas population of river Ken showed higher length increment.The results of the present study corroborated the previous findings of Sarkar et al. (2006).
The growth pattern in L. rohita showed increase in the populations of river Betwa, Sharda, Ganga and Gomti at certain age classes, indicating that the fish might have been exposed to favorable environmental conditions at these stages of life.In rivers Betwa and Sharda, the fishes were recorded up to 8+ ages whereas up to 7+ age recorded in Ganga and Gomti, which are quite closer to the maximum reported age of 10 years (Khan & Jhingran, 1975).The higher growth performance index in river Ken might be due to low competition, high productivity and better ecological conditions (Lakra, Sarkar, Dubey, Sani & Pandey, 2011).Sarkar et al. (2006) reported the analysis of the age structure of L. rohita in river Ganga and indicated that the number of older individuals tended to decrease which might be due to unsustainable exploitation of the resources, and the migration of the fishes away from the reaches that eventually dry out in order to find refuge in deeper habitats (Oliveira, Ferreira & Ferreira, 2002).The growth constant is important to understand the periods of life in the life span of fish.In L. rohita three phases of life were observed based on growth constant which also corroborates the studies of Singh (1990) who reported three phases of life in Indian major carps from riverine population of Aligarh and Jaisamand Lake, India.According to Yablokov (1986), the average growth characteristics can be used to separate populations of the same species or conspecific population.The present observations revealed that the wild populations of L. rohita showed variation in growth from one population to another, and fisheries strategy could be made accordingly for conservation and sustainable utilization.
The information derived from this study can be used to describe stock boundaries at a range of spatial scales that may assist in directing future studies to refine stock structure using other techniques.This study showed a well-defined growth pattern from different geographical locations and important information has been generated on the age pattern of L. rohita.It is evident that wild populations of L. rohita showed variations in age profile from one population to another.Therefore, special conservation measures may be taken up to improve the current situation and steps should be taken by the conservation agencies towards the stock enhancement of higher age classes.The present study will therefore be useful for the fishery biologists and conservation agencies

ACKNOWLEDGMENTS
The authors are grateful to the Director, National Bureau of Fish Genetic Resources, Lucknow, India for providing necessary facilities.Thanks are due to the Indian Council of Agricultural Research, New Delhi for financial support to carry out this work.Authors are also indebted to Daisy Arroyo Mora, Administrative Editor and two anonymous reviewers for constructive comments on this manuscript.

Fig. 1 .
Fig. 1.Map showing the sampling locations of Labeo rohita on different rivers of Ganga basin.

Fig. 5 .
Fig. 5. A., B., C. variation in different growth parameters of Labeo rohita; C l : specific rate of linear growth; C W : specific rate of weight increase; C lt : growth constant.

Fig. 6 .
Fig. 6.Percentage of age frequency of Labeo rohita in six drainages of Ganga basin in India.

TABLE 1
Sample size and GPS coordinates of Labeo rohita samplings sites and the implicating factors, Ganga basin, India Growth rate of L. rohita was analyzed by back-calculation methods, and the back-calculated lengths (L n ) of each age from the different populations are presented in Table2.From our results, we established up to a total of 8+ age class (the age class is the maximum length attainment in certain age group as detected through the back calculation method).In this study, a maximum 8+ age class was recorded from river Betwa and Sharda.The back-calculated length for 1+ age class was highest in river Sharda and lowest in Ganga, and for 8+ age class it was highest in river Scales of Labeo rohita showing different age classes from six rivers of Ganga basin in India.Comparative pattern of growth rate of L. rohita from six different rivers of Ganga basin, India the focus.Marks presumed to be annuli were counted under the profile projector.Age and growth: Fig. 2.