J. Appl. Ichthyol.

20 (2004), 407–412 Received: March 3, 2004

 2004 Blackwell Verlag, Berlin Accepted: June 9, 2004
ISSN 0175–8659

Randomly amplified polymorphic DNA analysis of four different populations of the

Indian major carp, Labeo rohita (Hamilton)
By M. S. Islam and M. S. Alam

Department of Fisheries Biology and Genetics, Bangladesh Agricultural University, Mymensingh, Bangladesh

Summary hatchery-produced seed are not performing in culture as well

The level of genetic variation provides the raw material for as in earlier years. Moreover, the mixing of hatchery-reared
selective improvement of a stock. Random amplified poly- rohu through escapes and the government’s massive seed
morphic DNA (RAPD) assay was used to assess the genetic stocking programme in open waterbodies may have resulted in
variation in three rivers: the Halda, the Jamuna and the Padma gene introgression into the pure wild stocks and, ultimately,
as well as in one hatchery population of the commercially may result in less well adapted fish in comparison with native
important Indian major carp, Labeo rohita. RAPD markers stock. It is thus essential to understand the genetic composition
were amplified from DNA samples of 35 fish from each of the of rohu for management of their natural populations in rivers.
four populations using six decamer random primers. The Genetic marker identification is especially needed for baseline
polymorphic loci proportions were 0.33, 0.28, 0.28 and 0.26 studies for monitoring potential changes in genetic makeup
and Nei’s gene diversity values were 0.06, 0.07, 0.06 and 0.05 and adaptive values as a result of interaction between wild and
for the Halda, the Jamuna, the Padma and the hatchery culture populations.
populations, respectively. The pairwise population differenti- Random amplified polymorphic DNA (RAPD) analysis is
ation (FST) values indicated a low level of genetic differenti- based on amplification of discrete regions of the genome by
ation between the population pairs. From the unweighted pair polymerase chain reaction (PCR) with short oligonucleotide
group method of arithmetic mean (UPGMA) dendrogram primers of arbitrary sequence (Welsh and McClelland, 1990;
based on Nei’s genetic distances a correlation between genetic Williams et al., 1990) to detect polymorphisms in the respect-
affinities and geographical area was found. The populations ive primer sites of the genome. Such polymorphisms inherit in
were segregated into two groups: the Halda in one group and a Mendelian fashion and can be used as genetic markers in
the Jamuna, the Padma and the hatchery in another group. discriminating different populations (Hadrys et al., 1992). The
Overall, the RAPD technique can be introduced as a tool in RAPD technique is relatively simple and inexpensive com-
the population genetics of the rohu fish to provide information pared with other methods (Ward and Grewe, 1995) as it does
on their genetic stock structure. not require prior DNA sequence knowledge for examination
of genomic variations.
In the present study, the RAPD method was applied to three
Introduction river and one hatchery population of rohu (L. rohita) to assess
Rohu, Labeo rohita (Cyprinidae: Cypriniformes), is naturally intraspecific genetic variation and relatedness among the
distributed in the rivers of Bangladesh, Burma, India and populations.
Pakistan (Jhingran and Pullin, 1985). In Bangladesh, this
species is mostly found in the Padma-Brahmaputra and Halda Materials and methods
river systems (Fig. 1). Among the world’s principal aqua-
Experimental fish
culture species, rohu production was ranked 7th
(754 677 tonnes) in 1998 (FAO, 2000; Hulata, 2001). Its good Rohu fry were obtained from fry traders as they were
taste and high market price makes it one of the favorite collecting from the three rivers, the Padma (6 mg), the Jamuna
freshwater aquaculture fishes in Bangladesh (23%) (BBS, (6 mg) and the Halda (5 mg). Fry were also collected from a
2000). In recent years, the natural breeding of rohu has become hatchery (5 mg) in the Jessore region (Fig. 1). The fry were
constrained by degradation of habitats as a result of environ- stocked in four 180 m2 earthen ponds (water depth: each 2 m;
mental modifications and anthropological intervention. The average water temperature: 28
C, range: 26–30
C; pH 7.93,
density of the species has declined significantly because of range: 7.07–9.11) for 2 months in the field laboratory of the
overfishing and lower recruitment over the past two decades. Faculty of Fisheries, Bangladesh Agricultural University,
River contribution as a natural source of major carp species Mymensingh. Initially, mustard oil cake dissolved in water
fry for aquaculture has been reduced to almost nil (1%) in was thrown into the ponds five times a day to feed the fry.
2003 as against 80% in the early 1980s (DoF, 2003). On the From the fifth day, supplemental feed consisting of rice bran,
contrary, the demand for fish fry has increased several- wheat bran and mustard oil cake in a 2 : 2 : 1 ratio (20% of
fold through aquaculture expansion; as a result, fry produc- body weight per day) was used. No water was exchanged
tion in public and private hatcheries has intensified. Little during the period of fry rearing. At the time of sampling on 15
attention, however, has been given to genetic quality main- September 2001, average body weight of the fish was 25 g.
tenance. Hybridization and inbreeding are very common Finally, to carry out the RAPD analysis, 35 individuals were
practices in Bangladesh hatcheries. As a consequence, the taken randomly from each population. A small piece of tissue

U.S. Copyright Clearance Centre Code Statement: 0175–8659/2004/2005–0407$15.00/0 www.blackwell-synergy.com

408 M. S. Islam and M. S. Alam

8.0). DNA was re-precipitated by adding two volumes of

absolute ethanol in the presence of 0.3 M sodium acetate, and
pelleted by centrifugation. The pellets were then washed with
70% ethanol, air-dried and re-suspended in an appropriate
volume of TE buffer. DNA quality was checked by electro-
phoresis in a minigel and quantified using a spectrophotometer
(Spectronic GenesisTM, Spectronic Instruments Inc., USA).

Primer selection
Initially, 45 decamer primers from three kits (20 from kit A, 20
from kit B and 5 from kit C) of random sequence (Operon
Technologies, Inc., Alameda, CA, USA) were screened on a
subsample of two randomly chosen fish from each population,
to test their suitability for amplifying rohu RAPDs that could
be accurately scored. Primers were evaluated on the basis of
intensity or resolution of bands, repeatability of markers and
consistency within individuals and potential to differentiate
populations (polymorphism). A final subset of six primers
(Table 1) of 45 exhibiting good quality banding patterns were
tested three times on subsamples to be certain that the bands
obtained were not mere artefacts of the RAPD method but
rather true amplified products; they were then selected for
analyses of the entire sample set of the four populations.

PCR amplification
The amplification conditions were based on Williams et al.
(1990), with some modifications. PCR reactions were per-
Fig. 1. Map of Bangladesh showing sampling sites (d) of Labeo formed on each DNA sample in a 10 ll reaction mix
rohita. Arrow indicates the point of confluence of the Jamuna and containing 1 ll of 10X Ampli Taq polymerase buffer, 2 ll of
Padma rivers 10 lM primer, 1 ll of 250 lM dNTPs (Takara, Japan), 1 unit
of Ampli Taq DNA polymerase (Takara) and 50 ng of
genomic DNA and a suitable amount of sterile deionized
was clipped with scissors from the caudal fin of each individual water. DNA amplification was performed in an oil-free
and immediately preserved in 95% ethanol. thermal cycler (Master Cycler Gradient, Eppendorf). The
reaction mix was preheated at 94
C for 3 min followed by 40
cycles of 1 min denaturation at 94
C, 1 min annealing at 34
C
Extraction of genomic DNA and elongation or extension at 72
C for 2 min. After the last
Genomic DNA was extracted from the fin tissue following cycle, a final step of 7 min at 72
C was added to allow
standard procedures (Alam et al., 1996). In brief, approxi- complete extension of all amplified fragments.
mately 30 mg of fin tissue was cut into small pieces, homo-
genized and digested with proteinase K in extraction buffer
[100 mM Tris-HCl, pH 8.0, 10 mM ethylenediaminetetraacetic Agarose gel electrophoresis
acid (EDTA) and 250 mM NaCl and 1% sodium dodecyl The amplified product from each sample was separated
sulphate (SDS)] overnight at 37
C. DNA was purified by electrophoretically on 1% agarose gel (Nacalai tesque, Inc.,
successive extraction with phenol : chloroform : isoamyl alco- Kyoto, Japan) containing ethidium bromide in 1X TAE buffer
hol (25 : 24 : 1, v/v/v) and chloroform : isoamyl alcohol at 120 V for 112 h (a 1.4% agarose gel was also tested but could
(24 : 1, v/v), respectively. DNA was precipitated first using not produce better resolution; thus 1% gel was used for all
0.6 volume of isopropanol, pelleted by centrifugation, then samples). A molecular weight marker DNA (Lambda DNA-
re-suspended in TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH EcoT14 I digest and/or 100 bp ladder) was electrophoresed

Table 1
Random amplified polymorphic DNA (RAPD) primers with corresponding bands scored and their size range together with polymorphic bands
observed in Labeo rohita

Number of Number of Size range

Primer codes Sequence (5¢–3¢) G + C (%) bands scored polymorphic bands (bp)

OPA04 AATCGGGCTG 60 10 0 230–1520

OPA09 GGGTAACGCC 70 4–9 5 280–1940
OPA13 CAGCACCCAC 70 6–7 1 400–1300
OPA18 AGGTGACCGT 60 7–9 2 210–1410
OPB12 CCTTGACGCA 60 3–8 6 420–1800
OPB13 TTCCCCCGCT 70 4–10 6 420–1850
RAPD analysis of Labeo rohita 409

alongside the RAPD reactions. DNA bands were observed on fragments scored for each individual, respectively (Lynch,
UV-transilluminator and photographed with a Gel Cam 1990). Within population similarity [Si] was calculated as the
Polaroid camera. average of SI across all possible comparisons between
individuals within a population. Between population similar-
ity (Sij) was calculated as the average similarity between
Data analysis randomly paired individuals from populations i and j (Lynch,
The RAPD markers were scored visually on the basis of their 1991).
presence (1) or absence (0), separately for each fish and each
primer. For more accuracy, band scoring was performed by
two independent persons. Bands not identified by the two Results
readers were considered as non-scorable. The scores obtained Of the six primers showing good technical resolution one
using all primers in the RAPD analysis were then pooled for primer (OPA04) resulted in monomorphic banding pattern
constructing a single data matrix. This was used for estima- and was excluded from analysis. Primer OPB13 generated the
ting polymorphic loci, Nei’s (1973) gene diversity, gene flow, highest number of bands whereas OPA13 produced the least
genetic distance and constricting an unweighted pair group number of bands. The five primers yielded a total of 43 distinct
method of arithmetic mean (UPGMA) dendrogram among bands of which 20 (46.5%) were polymorphic (either occurring
populations using the POPGENE (version 1.31) (Yeh et al., in or absent in <95% of all individuals) and therefore used in
1999) computer program. The same program was also used to the analysis. Among the five primers, OPB12 and OPB13
perform a test of homogeneity in different locus between resulted in the maximum number of polymorphic bands and
population pairs. Population differentiation (theta P or FST) thus a high level of polymorphism (Table 1). The RAPD
was estimated on the basis of square root transformation of profiles of different primers are shown in Fig. 2.
the frequencies of the null allele (recessive) genotype, and the
95% confidence interval (CI) was calculated from bootstrap-
ping over loci upon 1000 replications using the software Within and between population similarity indices
ÔTools for Population Genetic AnalysesÕ (TFPGA; Miller, Intrapopulation SI for the hatchery population was the highest
1997). (96.9%) followed by that of the Padma (94.8%), the Jamuna
The similarity index values (SI) between the RAPD profiles (94.3%) and the Halda (93.5%) river populations, respectively.
of any two individuals on the same gel were calculated from On the contrary, interpopulation SI (Sij) for the Padma River
RAPD markers according to the following formula: vs hatchery samples was higher (94.4%) than those for all
other between-population comparisons (Halda–Jamuna:
Similarity indexðSIÞ ¼ 2NAB =ðNA þ NB Þ
92.8%; Halda–Padma: 91.6%; Padma–Jamuna: 93.9%; Hal-
Where, NAB is the total number of RAPD bands shared by da–Hatchery: 92.3%; Jamuna–Hatchery: 93.8%). The Padma
individuals A and B, and NA and NB are the number of population was closer to the Jamuna population with a genetic

Fig. 2. Random amplified polymorphic DNA (RAPD) profiles of Labeo rohita for the primer OPA-09 (a), OPA-13 (b), OPA-18 (c), OPB-12 (d)
and OPB-13 (e) of the Operon kits A and B (Operon Technologies). M is the molecular weight marker (Lambda DNA-EcoT14 I digest and
100 bp DNA ladder)
410 M. S. Islam and M. S. Alam

Table 2 Table 4
Estimates of genetic variation: number and proportion of polymorphic Population differentiation (h or FST) based on square root transfor-
bands and gene diversity obtained in different populations of Labeo mation of frequencies of the null allele (recessive) genotype followed in
rohita parentheses by the 95% confidence interval (CI), gene flow (Nm) and
genetic distance (D) values between populations
Parameters Halda Jamuna Padma Hatchery
Gene Genetic
Total polymorphic bands 14 12 12 11 flow distance
Proportion 0.33 0.28 0.28 0.26 Populations h (FST) ± SD (95% CI) (Nm) (D)
of polymorphic bands
Gene diversity 0.06 0.07 0.06 0.05 Halda–Jamuna 0.043 ± 0.03 16.15 0.004
()0.005–0.105)
Number of overall polymorphic bands across populations: 20.
Halda–Padma 0.097 ± 0.087 08.34 0.008
Proportion of overall polymorphic loci across populations: 0.47.
()0.0086–0.252)
Overall gene diversity for all loci: 0.07.
Halda–Hatchery 0.078 ± 0.071 10.06 0.006
()0.008–0.200)
Jamuna–Padma 0.035 ± 0.030 19.01 0.004
()0.009–0.085)
Jamuna–Hatchery 0.038 ± 0.028 17.61 0.004
Table 3 ()0.004–0.089)
Chi-square values for the loci causing significant departure from Padma–Hatchery 0.018 ± 0.019 26.19 0.002
homogeneity in different population pairs (below diagonal) of Labeo ()0.005–0.071)
rohita All populations 0.0508 ± 0.0350 9.34 –
(0.00132–0.1178)
Loci Halda Jamuna Padma

Jamuna
OPB12-8 6.34**
Padma Halda
OPA13-1 5.36*
OPB12-8 17.03***
OPA09-1 5.43* Jamuna
Hatchery
OPB12-6 6.03**
OPB12-8 11.71***
OPA09-1 4.57* Padma
OPB12-6 6.03** 6.03**
OPA13-1 4.15*
Hatchery
*P < 0.05; **P < 0.01; ***P < 0.001.

0.008 0.006 0.004 0.002 0.000

similarity of 93.9% when compared with the Halda–Jamuna Genetic distance
and the Halda–Padma populations. Fig. 3. Unweighted pair group method of arithmetic mean (UPGMA)
dendrogram based on Nei’s (1972) genetic distance, summarizing data
on differentiation between Labeo rohita populations according to
Polymorphism in different populations random amplified polymorphic DNA (RAPD) analysis
The highest (0.33) and lowest (0.26) frequencies of polymor-
phic bands across all five primers were observed in the Halda
River and the hatchery population, respectively (Table 2). The population pairs (Table 4); however, none of the pairwise
Jamuna (0.28) and the Padma (0.28) river populations showed FST values was found to be significant. The estimated gene
similar polymorphism in the RAPD loci. Among the poly- flow (Nm) value across the populations was 9.34. The
morphic loci four (OPA09-1, OPA13-1, OPB12-6 and OPB12- highest Nm value (26.19) was observed between the Padma
8) were found to cause significant departure from homogeneity and the hatchery populations, whereas the lowest Nm (8.34)
in the population pairs (Table 3). value was found between the Halda and the Padma
populations (Table 4). Among the natural stocks a relatively
high Nm value (19.01) was estimated between the Padma
Nei’s gene diversity and the Jamuna populations.
Overall gene diversity across all populations for all loci studied
was 0.07. The Jamuna River population showed relatively
higher gene diversity compared with the Padma and the Halda Genetic distance
populations (Table 2). The gene diversity value for the The genetic distance value for the Halda–Padma population
hatchery population was the lowest. However, the 95% pair was the highest (0.008) whereas the value for the Padma–
confidence intervals showed no significant differences in gene Hatchery population pair was the lowest (0.002) (Table 4). The
diversity between the populations. UPGMA dendrogram based on Nei’s (1972) genetic distance
indicated the segregation of four populations of rohu into two
clusters: the Halda alone belonged to one cluster whereas the
Population differentiation and gene flow Jamuna, Padma and hatchery populations made another
The FST value between the Halda and the Padma popula- cluster. The second cluster was further separated into two
tions was the highest and the FST value between the Padma subgroups: the Jamuna in one group and the Padma and the
and the hatchery populations was the lowest among all hatchery in a separate group (Fig. 3).
RAPD analysis of Labeo rohita 411

Discussion polymorphisms (5–10% of the bands were polymorphic) for

Despite the importance of rohu as a foodfish in the Indian both channel catfish (Ictalurus punctatus) and blue catfish
subcontinent, information on the genetic background of this (I. furcatus) strains using 100 RAPD primers, although distinct
species is very scarce. Allozyme analysis of 20 individuals per phenotypes were observed and inherited in each of these
population using four enzyme systems encoding six gene loci strains (Dunham et al., 1993). In the present study, 46.5% of
revealed similar polymorphism in a river and a hatchery stock the bands were found to be polymorphic, much higher than
of rohu, although the overall genetic variation was slightly that obtained by Liu et al. (1999) in catfish and indicating that
higher in the river population (Alam et al., 2002). Simonsen the RAPD system may be more useful to generate molecular
et al. (2004) observed a high incidence of hybridization in the markers for genetic characterization in the Indian major carp,
hatchery stocks of the three Indian major carps (Catla catla, L. rohita.
L. rohita, and Cirrhinus mrigala) by isozyme electrophoresis.
The present study was the first attempt to determine the
Acknowledgements
genetic variation among different populations of rohu using
RAPD markers. The research was supported by a grant of the International
In this study, the lowest genetic variability was found within Foundation for Science (grant no. A/3103). V. Simonsen,
the hatchery population. This may be attributed to the National Environmental Research Institute, Denmark is
maintenance of a limited number of individuals sampled from acknowledged for reading the manuscript and making com-
the wild and their repeated propagation over a long period. ments for its improvement. Authors are also grateful to the
Inbreeding may be another reason for reduced genetic variation two unknown reviewers who critically reviewed the manuscript
in the hatchery population, as was also reported by Eknath and and gave suggestions for improvement.
Doyle (1990) on the basis of effective population size. Barman
et al. (2003) also observed a high level of within-species genetic
References
similarity (97.5%), i.e. low levels of within-species genetic
Alam, G. N., 2003: Natural carp breeding-ground under threat. The
variation in farmed stocks of major carps in India.
Daily Star Web Edition, Vol. 4, August 8, 2003.
The higher similarities and lower genetic differences and the Alam, M. S.; Popplewell, A.; Maclean, N., 1996: Germline transmis-
higher level of gene flow between the Padma River and sion and expression of a lacZ containing transgene in tilapia
hatchery populations and between the Jamuna River and (Oreochromis niloticus). Transgenic Res. 5, 87–95.
hatchery populations indicate that the hatchery stock might Alam, M. A.; Akanda, M. S. H.; Khan, M. M. R.; Alam, M. S., 2002:
Comparison of genetic variability between a hatchery and a river
have originated from one or both of the rivers. As the Padma population of rohu (Labeo rohita) by allozyme electrophoresis.
and Jamuna rivers join near the Goalanda upazilla of Rajbari Pak. J. Biol. Sci. 5, 959–961.
district (Fig. 1), a possible mixing between the two populations Azadi, M. A.; Kibria, M. M.; Jahangir, S.; Akhteruzzaman, M., 2003:
cannot be excluded. The higher genetic distance between the Management of spawn fishery of major carps (Catla catla, Labeo
rohita and Cirrhinus mrigala) in the Halda River, Bangladesh.
Halda and the Jamuna or between the Halda and the Padma
http://www.lars2.org/unedited_papers/Azadi.pdf.
populations may be explained by the presence of a geograph- Barman, H. K.; Barat, A.; Yadav, B. M.; Banerjee, S.; Meher, P. K.;
ical barrier between each pair. Reddy, P. V. G. K.; Jana, R. K., 2003: Genetic variation between
Bandsharing-based similarity indices were marginally higher four species of Indian major carps as revealed by random
for intrapopulation samples (average 94.9) than for all amplified polymorphic DNA assay. Aquaculture 217, 115–123.
BBS, 2000: Bangladesh Bureau of Statistics. Species composition of
comparisons between population samples (average 93.1). This pond catches, 1999–2000. Statistical Yearbook of Fishes, Depart-
implies that individuals within each population are genetically ment of Fisheries, Government of Bangladesh.
more similar to each other, as is expected, than to individuals DoF, 2003: Fish-fortnight compendium. Department of Fisheries,
from all other populations. The populations, however, were Ministry of Fisheries and Livestock, Peoples Republic of Bangladesh.
Dunham, R. A.; Hyde, C.; Masser, M.; Smitherman, R. O.; Perez, R.;
very similar to each other genetically.
Plumb, J.; Ramboux, A. C., 1993: Comparison of culture traits of
The proportion of polymorphic loci in the Halda population channel catfish, Ictalurus punctatus, and blue catfish, I. furcatus.
(0.33) is indicative of a relatively high level of genetic variation. J. Appl. Aquacult. 3, 257–267.
Additionally, four RAPD markers were found to be specific to Eknath, A. E.; Doyle, R. W., 1990: Effective population size and rate
the Halda population that was absent in the other three of inbreeding in aquaculture of Indian major carps. Aquaculture
85, 293–305.
populations (results not shown). The Halda is a geographically FAO, 2000: World aquaculture production by principal species in
isolated freshwater tidal river originating in the hilly region of 1998. FAO, Rome, Italy.
eastern Bangladesh. No spawning grounds of major carps were Hadrys, H.; Balick, M.; Schierwater, B., 1992: Application of random
reported within the freshwater tidal zone in the region except in amplified polymorphic DNA (RAPD) in molecular ecology. Mol.
Ecol. 1, 55–63.
the Halda (Azadi et al., 2003) and therefore no chance of mixing
Hulata, G., 2001: Genetic manipulation in aquaculture: a review of
with other stocks. Overall genetic variation in the three river stock improvement by classical and modern technologies. Gen-
populations appears to be low, with the reasons difficult to etics 111, 155–173.
explain. However, in the Halda it may be explained by its Jhingran, V. G.; Pullin, R. S. V., 1985: A hatchery manual for the
smaller size (only 35 km length) and declining fish populations: common, Chinese and Indian major carps. ICLARM studies and
reviews 11. Asian Development Bank, Manila, Philippines and
the number of breeding fish is decreasing at a rate of 25–30% ICLARM, Manila, Philippines, 191 pp.
per year because of wanton capture of broodfish during the Liu, Z. J.; Li, P.; Argue, B. J.; Dunham, R. A., 1999: Random
breeding period (April–May) by nearby residents (Alam, 2003). amplified polymorphic DNA markers: usefulness for gene map-
Significant departure from homogeneity only at four loci (of ping and analysis of genetic variation of catfish. Aquaculture 174,
59–68.
43) and insignificant population differentiation as per the FST
Lynch, M., 1990: The similarity index and DNA fingerprinting. Mol.
values across all loci indicate a relatively low level of genetic Biol. Evol. 7, 478–484.
differentiation between the studied river and hatchery stocks. Lynch, M., 1991: Analysis of population genetic structure by
Liu et al. (1999) observed that very low levels of intraspecific DNA fingerprinting. In: DNA fingerprinting approaches and
412 M. S. Islam and M. S. Alam

applications. T. Burke, G. Dolf, A. J. Jeffreys and R. Wolf (Eds). Welsh, J.; McClelland, M., 1990: Fingerprinting genomes using PCR
Basel, Switzerland, pp. 113–126. with arbitrary primers. Nucleic Acids Res. 18, 7213–7218.
Miller, M. P., 1997: Tools for Population Genetic Analyses, TFPGA, Williams, J. G. K.; Kubelik, A. R.; Livak, K. J.; Rafalski, J. A.;
version 1.3. Computer software distributed by the author Tingey, S. V., 1990: DNA polymorphisms amplified by arbitrary
([email protected]). primers are useful as genetic markers. Nucleic Acids Res. 18,
Nei, M., 1972: Genetic distance between populations. Am. Nat. 106, 6531–6535.
283–292. Yeh, F. C.; Yang, R. C.; Boyle, T., 1999: POPGENE version 1.31:
Nei, M., 1973: Analysis of gene diversity in subdivided populations. Microsoft Window-based freeware for population genetic analysis
Proc. Natl Acad. Sci. U S A 70, 3321–3323. (http://www.ualberta.ca/~fyeh).
Simonsen, V.; Hansen, M. M.; Sarder, M. R. I.; Alam, M. S., 2004:
High level hybridization in three species of Indian major carps. Author’s address: Md. Samsul Alam, Department of Fisheries Biology
NAGA 27 (in press). and Genetics, Bangladesh Agricultural University,
Ward, R. D.; Grewe, P. M., 1995: Appraisal of molecular techniques in Mymensingh 2202, Bangladesh.
fisheries. In: Molecular genetics in fisheries. G. R. Carvalho and E-mail: [email protected]
T. J. Pitcher (Eds). Chapman and Hall, London, pp. 29–54.