GENETIC VARIABILITY IN SOME COMMERCIALLY IMPORTANT CARNIVOROUS FISHES BY MOLECULAR MARKERS IN PUNJAB-PAKISTAN

By Ehsan Mehmood Bhatti

2007-GCUF-2357-226

Thesis submitted in partial fulfillment of

the requirement for the degree of

DOCTOR OF PHILOSOPHY

IN

ZOOLOGY

DEPARTMENT OF WILDLIFE AND FISHERIES

GC UNIVERSITY FAISALABAD

2012

ii

DECLARATION CERTIFICATE BY THE RESEARCH SUPERVISOR

I certify that the contents and form of thesis submitted by Mr. Ehsan Mehmood Bhatti, Registration No. 2007-GCUF-2357-226 has been found satisfactory and in accordance with the prescribed format. I recommend it to be processed for the evaluation by the External Examiner for the award of degree.

Signature………………….

Prof. Dr. Naureen Aziz Qureshi

Designation with Stamp………………….

Chairperson

Signature………………….

Prof. Dr. Naureen Aziz Qureshi

Designation with Stamp………………….

Dean / Academic Coordinator

Signature………………….

Prof. Dr. Naureen Aziz Qureshi

iii

SUPERVISORY COMMITTEE SUPERVISOR Prof. Dr. Naureen Aziz Qureshi Professor ______Department of Wildlife and Fisheries

Member-1 Dr. Muhammad Siddique Assistant Professor ______Department of Environmental Sciences

Member-2 Dr. Hafiz Abubakar Saddiqi Assistant Professor ______Department of Wildlife and Fisheries

SCRUTINIZING COMMITTEE

Member-1 Member-2 Signature…………………. Signature………………… Dr. Muhammad Mushtaq-ul-hasan Dr. Hafiz Abubakar Saddiqi Associate professor Assistant Professor Department of Wildlife and Fisheries Department of Wildlife and Fisheries

Member-3 Member-4 Signature…………………. Signature…………………. Dr. Farhat Jabeen Dr. Sajid Yaqub Qureshi Assistant Professor Assistant Professor Department of Wildlife and Fisheries Department of Wildlife and Fisheries Member-5 Signature…………………. Dr. Tyyaba Ali Lecturer Department of Wildlife and Fisheries

iv

LIST OF TABLES Table 1 Sequences of the used primers ...... 157 Table 2 ANOVA for the body weight of Channa punctatus ...... 162 Table 3 ANOVA for the Total Length of C. punctatus ...... 163 Table 4 ANOVA for the Head Length of C. punctatus ...... 163 Table 5 ANOVA for the Stoutness of C. punctatus ...... 164 Table 6 ANOVA for the Dorsal Fin Length of C. punctatus ...... 165 Table 7 ANOVA for the Caudal Fin Length of C. punctatus ...... 165 Table 8 ANOVA for the Average Length of Paired Pectoral Fins of C. punctatus .... 166 Table 9 ANOVA for the Wet Body Weight of Channa marulius ...... 167 Table 10 ANOVA for the Total Length of C. marulius ...... 168 Table 11 ANOVA for the Head Length of C. marulius ...... 168 Table 12 ANOVA for the Stoutness of C. marulius ...... 169 Table 13 ANOVA for the Dorsal Fin Length of C. marulius ...... 170 Table 14 ANOVA for the Caudal Fin Length of C. marulius ...... 170 Table 15 ANOVA for the Anal Fin Length of C. marulius ...... 171 Table 16 ANOVA for the Average Length of Paired Pectoral Fins of C. marulius .... 172 Table 17 ANOVA for the Wet Body Weight of Rita rita ...... 173 Table 18 ANOVA for the Fork Length of R. rita ...... 173 Table 19 ANOVA for the Total Length of R. rita ...... 174 Table 20 ANOVA for the Head Length of R. rita ...... 175 Table 21 ANOVA for the Stoutness of R. rita ...... 175 Table 22 ANOVA for the Dorsal Fin Length of R. rita ...... 176 Table 23 ANOVA for the Caudal Fin Length of R. rita ...... 177 Table 24 ANOVA for the Anal Fin Length of R. rita ...... 177 Table 25 ANOVA for the Adipose Fin Length of R. rita ...... 178 Table 26 ANOVA for the Average Length of Paired Pectoral Fins of R. rita ...... 179 Table 27 ANOVA for the Average Length of Paired Pelvic Fins of R. rita ...... 180 Table 28 ANOVA for the Wet Body Weight of Sperata seenghala ...... 181 Table 29 Analysis of variance on the Fork Length of S. seenghala ...... 181 Table 30 ANOVA for the Total Length of S. seenghala ...... 182 Table 31 ANOVA for the Head Length of S. seenghala ...... 183 Table 32 ANOVA for the Stoutness of S. seenghala ...... 183 Table 33 ANOVA for the Dorsal Fin Length of S. seenghala ...... 184 Table 34 ANOVA for the Caudal Fin Length of S. seenghala ...... 185 Table 35 ANOVA for the Anal Fin Length of S. seenghala ...... 186 Table 36 ANOVA for the Adipose Fin Length of S. seenghala ...... 187 Table 37 ANOVA for the Av. Length of Paired Pectoral Fins of S. seenghala ...... 187 Table 38 ANOVA for the Av. Length of Paired Pelvic Fins of S. seenghala ...... 188 Table 39 ANOVA for the Wet Body Weight of Wallago attu ...... 189 Table 40 ANOVA for the Fork Length of W. attu ...... 190 Table 41 ANOVA for the Total Length of W. attu ...... 191 Table 42 ANOVA for the Head Length of W. attu ...... 191 Table 43 ANOVA for the Stoutness of W. attu ...... 192 Table 44 ANOVA for the Dorsal Fin Length of W. attu ...... 193 Table 45 ANOVA for the Caudal Fin Length of W. attu ...... 193 Table 46 ANOVA for the Average Length of Paired Pectoral Fins of W. attu ...... 194 Table 47 ANOVA for the Average Length of Paired Pelvic Fins of W. attu ...... 195 Table 48 Summary statistics of PCA for C. punctatus ...... 197

v

Table 49 Correlation matrix (Pearson (n)) of PCA for C. punctatus ...... 197 Table 50 Bartlett's sphericity test of PCA for C. punctatus ...... 198 Table 51 Eigen values of PCA for C. punctatus ...... 198 Table 52 Eigen vectors of PCA for C. punctatus ...... 200 Table 53 Factor loadings of PCA for C. punctatus ...... 200 Table 54 Correlations between variables and factors of PCA for C. punctatus ...... 201 Table 55 Contribution of the variables (%) of PCA for C. punctatus ...... 203 Table 56 Squared cosines of the variables of PCA for C. punctatus ...... 203 Table 57 Summary statistics for C. marulius ...... 208 Table 58 Correlation matrix (Pearson (n)) for C. marulius ...... 208 Table 59 Bartlett’s sphericity test for C. marulius ...... 209 Table 60 Eigen values of PCA for C. marulius ...... 209 Table 61 Eigen vectors of PCA for C. marulius ...... 210 Table 62 Factor loadings of PCA for C. marulius ...... 211 Table 63 Correlations between variables and factors of PCA for C. marulius ...... 212 Table 64 Contribution of the variables (%) of PCA for C. marulius ...... 214 Table 65 Squared cosines of the variables of PCA for C. marulius ...... 214 Table 66 Summary statistics for R. rita ...... 219 Table 67 Correlation matrix (Pearson (n)) for R. rita ...... 220 Table 68 Bartlett's sphericity test for R. rita ...... 220 Table 69 Eigen values of Principal Component Analysis f (PCA) for R. rita ...... 221 Table 70 Eigen vectors of PCA for R. rita ...... 223 Table 71 Factor loadings of PCA for R. rita ...... 223 Table 72 Correlations between variables and factors of PCA for R. rita ...... 224 Table 73 Contribution of the variables (%) of PCA for R. rita ...... 226 Table 74 Squared cosines of the variables of PCA for R. rita ...... 226 Table 75 Summary statistics for S. seenghala ...... 231 Table 76 Correlation matrix (Pearson (n)) for S. seenghala ...... 232 Table 77 Bartlett's sphericity test for S. seenghala ...... 232 Table 78 Eigen values of PCA for S. seenghala ...... 233 Table 79 Eigen vectors of PCA for S. seenghala ...... 234 Table 80 Factor loadings of PCA for S. seenghala ...... 235 Table 81 Correlations between variables and factors of PCA for S. seenghala ...... 236 Table 82 Contribution of the variables (%) of PCA for S. seenghala ...... 237 Table 83 Squared cosines of the variables of PCA for S. seenghala ...... 238 Table 84 Summary statistics for W. attu ...... 241 Table 85 Correlation matrix (Pearson (n)) for W. attu ...... 242 Table 86 Bartlett's sphericity test for W. attu ...... 243 Table 87 Eigen values of PCA for W. attu ...... 243 Table 88 Eigen vectors of PCA for W. attu ...... 244 Table 89 Factor loadings of PCA for W. attu ...... 245 Table 90 Correlations between variables and factors of PCA for W. attu ...... 246 Table 91 Contribution of the variables (%) of PCA for W. attu ...... 247 Table 92 Squared cosines of the variables of PCA for W. attu ...... 248 Table 93 Average Physicochemical Parameters of the Study Sites ...... 250 Table 94 Correlation Matrix of Physico-chemical Parameters of the Study Sites .... 253 Table 95 Primers used and amplification data for Channa punctatus ...... 255 Table 96 Primers used and amplification data for Channa marulius ...... 256 Table 97 Primers used and amplification data for Rita rita ...... 258 Table 98 Primers used and amplification data for Sperata seenghala ...... 259

vi

Table 99 Primers used and amplification data for Wallago attu ...... 260 Table 100 Variance decomposition for the optimal classification of C. punctatus .... 275 Table 101 Distances between the class centroids of C. punctatus ...... 275 Table 102 Central Objects of the Classes of C. punctatus and Distances between them ...... 276 Table 103 Results by class of C. punctatus ...... 276 Table 104 Variance decomposition for the optimal classification of C. marulius ...... 283 Table 105 Distances between the class centroids of C. marulius ...... 283 Table 106 Central Objects of the Classes of C. marulius and Distances between them ...... 284 Table 107 Results by class of C. marulius ...... 284 Table 108 Variance decomposition for the optimal classification of Rita rita ...... 291 Table 109 Distances between the class centroids of R. rita ...... 291 Table 110 Central Objects of the Classes of R. rita and Distances between them .. 292 Table 111 Results by class of Rita rita ...... 293 Table 112 Variance decomposition for the optimal classification of S. seenghala ... 299 Table 113 Distances between the class centroids of S. seenghala ...... 299 Table 114 Central Objects of the Classes of S. seenghala and Distances ...... 299 Table 115 Results by class of S. seenghala ...... 299 Table 115 Variance decomposition for the optimal classification of W. attu ...... 306 Table 116 Distances between the class centroids of W. attu ...... 306 Table 117 Central Objects of the Classes of W. attu and Distances between them 307 Table 118 Results by class of W. attu ...... 307 Table 119 Eigen values of RAPD data (PCA) for C. punctatus ...... 309 Table 120 Rotation matrix for C. punctatus ...... 314 Table 121 Percentage of variance after Varimax rotation for C. punctatus ...... 314 Table 122 Eigen values of RAPD data (PCA) for C. marulius ...... 319 Table 123 Rotation matrix after Varimax rotation for C. marulius ...... 325 Table 124Percentage of variance after Varimax rotation C. marulius ...... 326 Table 125 Eigen values of RAPD data (PCA) for Rita rita ...... 331 Table 126 Rotation matrix for R. rita ...... 336 Table 127 Percentage of variance after Varimax rotation for R. rita ...... 336 Table 128 Eigen values of RAPD data (PCA) for S. seenghala ...... 340 Table 129 Rotation matrix for S. seenghala ...... 345 Table 130 Percentage of variance after Varimax rotation for S. seenghala ...... 345 Table 131 Eigen values of RAPD data (PCA) for W. attu ...... 350 Table 132 Rotation matrix for W. attu ...... 355 Table 133 Percentage of variance after Varimax rotation for W. attu ...... 355

vii

LIST OF FIGURES Figure 1 Map showing the sampling sites at different Rivers in Punjab-Pakistan .... 149 Figure 2 Representative picture of the target species Channa punctatus (Daula) ... 149 Figure 3 Representative picture of the target species Channa marulius (Soal) ...... 150 Figure 4 Representative picture of the target species Rita rita (Khagga) ...... 150 Figure 5 Representative picture of Sperata seenghala (Singhara)...... 151 Figure 6 Representative picture of the target species Wallago attu (Mulley) ...... 151 Figure 7 Scree plot between Eigen values, factors and cumulative variability for C. punctatus ...... 199 Figure 8 Graph between variables for C. punctatus ...... 202 Figure 9 Graph between Observations for C. punctatus ...... 204 Figure 10 Biplot graph between factors for C. punctatus ...... 205 Figure 11 Scree plot between Eigen values, factors and cumulative variability for C. marulius ...... 210 Figure 12 Graph between variables for C. marulius ...... 213 Figure 13 Graph between Observations for C. marulius ...... 215 Figure 14 Biplot graph between factors for C. marulius ...... 216 Figure 15 Scree plot between Eigen values, factors and cumulative variability for R. rita ...... 221 Figure 16 Graph between variables for R. rita ...... 225 Figure 17 Graph between Observations for R. rita ...... 227 Figure 18 Biplot graph between factors for R. rita ...... 228 Figure 19 Scree plot between Eigen values, factors and cumulative variability for S. seenghala ...... 233 Figure 20 Graph between variables for S. seenghala ...... 236 Figure 21 Graph between Observations for S. seenghala ...... 239 Figure 22 Biplot graph between factors for S. seenghala...... 240 Figure 23 Scree plot between Eigen values, factors and cumulative variability for W. attu ...... 243 Figure 24 Graph between variables for W. attu ...... 247 Figure 25 Graph between Observations for W. attu ...... 249 Figure 26 Biplot graph between factors for W. attu ...... 250 Figure 27 Picture showing the Amplification of the OPB-2 for R. rita (1-4) and S. seenghala (5-8) for the samples collected from Baloki ...... 262 Figure 28 Picture showing the Amplification of OPD-4 for W. attu for the samples collected from Baloki ...... 263 Figure 29 Picture showing the Amplification of OPD-4 for R. rita (1-4) and S. seenghala (5-8) for the samples collected from Qadirabad ...... 264 Figure 30 Picture showing the Amplification of OPC-11 for C. punctatus (1-3) and C. marulius (4-7) for the samples collected from Baloki, River Ravi...... 265 Figure 31 Picture showing the Amplification of OPD-5 for C. marulius (1-4) and C. punctatus (5-8) for collected samples from Trimu and Taunsa barrages ...... 266 Figure 32 Picture showing the Amplification of OPD-1 for S. seenghala (1-5) and OPD-4 for R. rita (6-10) from Taunsa barrage...... 267 Figure 33 Picture showing the Amplification of OPB-2 for W. attu for samples collected from Trimu and Taunsa Barrages...... 268 Figure 34 Picture showing the amplification of OPC-11 and OPC-15 for S. seenghala for samples from Qadirabad and Chashma barrages ...... 269 Figure 35 Dendrogram showing classification of Channa punctatus...... 271

viii

Figure 36 Dendrogram showing classes of C. punctatus...... 273 Figure 37 Dendrogram showing classification of Channa marulius...... 279 Figure 38 Dendrogram showing classes of C. marulius...... 281 Figure 39 Dendrogram showing classification of R. rita...... 287 Figure 40 Dendrogram showing classes of R. rita...... 289 Figure 41 Dendrogram showing classification of S. seenghala...... 295 Figure 42 Dendrogram showing classes of S. seenghala...... 297 Figure 43 Dendrogram showing classification of W. attu...... 302 Figure 44 Dendrogram showing classes of W. attu...... 304 Figure 45 Graph between Eigen values and cumulative variability for Channa punctatus ...... 310 Figure 46 Graph between variables and Factors for C. punctatus ...... 310 Figure 47 Graph between Observations and factors for C. punctatus ...... 311 Figure 48 Biplot graph between F factors for C. punctatus ...... 312 Figure 49 Graph between variables after Varimax rotation for C. punctatus ...... 315 Figure 50 Graph between observations after Varimax rotation for C. punctatus ...... 316 Figure 51 Biplot graph between F factors after Varimax rotation for C. punctatus ... 317 Figure 52 Graph between Eigen values and cumulative variability for C. marulius . 320 Figure 53 Graph between variables and Factors for C. marulius ...... 321 Figure 54 Graph between Observations and factors for C. marulius ...... 322 Figure 55 Biplot graph between F factors for C. marulius ...... 323 Figure 56 Graph between variables after Varimax rotation for C. marulius...... 326 Figure 57 Graph between observations after Varimax rotation for C. marulius ...... 327 Figure 58 Biplot graph between F factors after Varimax rotation for C. marulius ..... 328 Figure 59 Graph between Eigen values and cumulative variability for R. rita ...... 331 Figure 60 Graph between variables and Factors for R. rita ...... 332 Figure 61 Graph between Observations and factors for R. rita ...... 333 Figure 62 Biplot graph between F factors for R. rita ...... 334 Figure 63 Graph between variables after Varimax rotation for R. rita ...... 337 Figure 64 Graph between observations after Varimax rotation for R. rita ...... 337 Figure 65 Biplot graph between F factors after Varimax rotation for R. rita ...... 338 Figure 66 Graph between Eigen values and cumulative variability for S. seenghala ...... 340 Figure 67 Graph between variables and Factors for S. seenghala ...... 341 Figure 68 Graph between Observations and factors for S. seenghala ...... 342 Figure 69 Biplot graph between F factors for S. seenghala ...... 343 Figure 70 Graph between variables after Varimax rotation for S. seenghala ...... 346 Figure 71 Graph between observations after Varimax rotation for S. seenghala ..... 347 Figure 72 Biplot graph between F factors after Varimax rotation for S. seenghala .. 348 Figure 73 Graph between Eigen values and cumulative variability for W. attu ...... 350 Figure 74 Graph between variables and Factors for W. attu ...... 351 Figure 75 Graph between Observations and factors for W. attu ...... 352 Figure 76 Biplot graph between F factors for W. attu ...... 353 Figure 77 Graph between variables after Varimax rotation for W. attu ...... 356 Figure 78 Graph between observations after Varimax rotation for W. attu ...... 357 Figure 79 Biplot graph between F factors after Varimax rotation for W. attu ...... 358

ix

TABELE OF CONTENTS

CHAPTERS TITLE PAGE A Declaration ii

B Certificate by the research supervisor iii

C List of tables vi

D List of Figures ix

E Table of contents xi

F Dedication xi

G Acknowledgement Xii

H Summary xv

1 Introduction 1

2 Review of literature 9

3 Materials and methods 30

4 Results 40

5 Discussion 173

6 References 197

7 Appendices 218

x

Dedication

DEDICATED

TO

HOLY PROPHET

HAZARAT MUHAMMAD (PEACE BE UPON HIM)

AND

My Loving Mother and (Late) Father

WHOSE

Prayers, love and guidance enlightened my whole life

My wife and children

Without their support I am nothing

xi

ACKNOWLEDGEMENTS

Allah never spoils any effort. I set my unfeigned and meek thanks before Him, Who created the universe and bestowed the mankind with knowledge and wisdom to search for its secrets and invigorated me with the fortitude and capability to complete my research work and enabled me to contribute a drop to the existing oceans of scientific knowledge.

But, the real success for any person in this world is that for which I have no words to pay in the court of Holly Prophet Muhammad (S.A.W) that I am in his Ummah, and his moral and spiritual teachings enlightened my heart, mind and flourished my thoughts towards achieving high ideas of life.

I am very thankful to, Prof. Dr. Naureen Aziz Qureshi, Dean Faculty of Sciences and Technology, Head of the Department of Wildlife and Fisheries, GC University Faisalabad, who gave me the opportunity of opting research work.

I am very grateful to my supervisor Prof. Dr. Naureen Aziz Qureshi, Professor Department of Wildlife and fisheries, Government College University Faisalabad, for her guidance, cooperation, constructive criticism, keen interest, intensive teaching and dynamic supervision throughout the course of studies and research endeavor.

I also acknowledge Mr. Muhammad Fayaz, a PhD research scholar for valuable guidance and facilitating my work, Dr. Hafiz Abubakar Saddiqi, Assistant Professor, Department of Wild life and Fisheries and Dr. Muhammad Siddique, Assistant Professor, Department of Environmental science for their cooperation, help, encouragement and guidance in conducting my experimental work.

My special thanks to Mr. Muhammad Zahid, who is working in the Zoology Research lab and he offered his cooperation, beside services in the lab without which my research work could never be completed

Friends are the comrades of the battle, the battle to generate knowledge, sift myths and facts and to remove ambiguity. They were co-sharer of my struggle and my work. I express my thankful feelings for my friends specially Mr. Zahid Ali,

xii

Dr. Zahid Sharif Mirza and field staff of the Fisheries Department for their sincere help and moral encouragement.

No acknowledgement could ever adequately express my obligations to my affectionate and adoring Mother and (Late) Father ( He died during my PhD studies) whose hands always raised in prayers for me and without whose moral and financial support, the present distinction would have merely been a dream. They always acted as a lighthouse for me in the dark oceans of life path. No words can really express the feelings that I have for my beloved family. They spend time without me even at times when they severely need me at home.

May Allah Almighty infuse me with the energy to fulfill their noble inspirations and expectations and further edify my competence (Amin).

EHSAN MEHMOOD BHATTI

xiii

Summary

To conserve the fish biodiversity, there is urgent need to study the genetic diversity of natural fish stocks for successful fishery management, conservation, and rehabilitation of the species. For this purpose, the present study was planned to evaluate the use of the RAPD assay to generate the species specific profile, to estimate the genetic similarity/variation and to examine the genomic variation based on RAPD data and the population genetic structure for conservation of five carnivorous fishes named Channa punctatus, Channa marulius, Sperata seenghala, Rita rita and Wallago attu. The fish samples for this study were collected from Chashma Barrage at River Indus near District Mianwali, Qadirabad Barrage at Chenab River in Tehsil Wazirabad, District Gujranwala, Baloki Barrage at Ravi River in Tehsil Bhai Pharo District Kasur, Trimu Barrage at the junction of Chenab and Jhelum Rivers in District Jhang and Taunsa Barrage at Indus River in Tehsil Kot Addu, District Muzaffar Garh.

Morphometric parameters of the collected fish specimens, were measured and the data was subjected to ANOVA and multivariate Principle Component Analysis (PCA). Physicochemical parameters of the water bodies such as Water Temperature, Electrical Conductivity, pH, Dissolved Oxygen, Salinity at the spot and Total Dissolved Solids (T.D.S.), Total Alkalinity, Total Hardness, were measured from the water samples brought to the lab and noted. Correlation was calculated with the help of computer. The DNA from these fish specimens was extracted by modified salt extraction method and its presence was qualitatively measured with the gel electrophoreses. PCR was performed in the lab. Binary data of the RAPD for different species were subjected to XLSTAT-2012, version 1.02 and Dendrogram were generated. The binary data was further analyzed for diversity indices and Principle Component Analysis (PCA).

The morphometric data of stoutness and average length of paired pectoral fin length in the ANOVA show significant difference in case of Channa punctatus. All studied parameters were significantly different in S. seenghala and non-significant in Channa marulius, Rita rita and Wallago attu. When the PCA was performed for the morphometric parameters, C. punctatus divided into four factors, first two main factors accounted for 98.706% of cumulative variability, in C. marulius, divided into

xiv six factors with first two main factors accounted for 99.996%, in tested variables of the R. rita division was into eight factors with two main components, which accounted for 99.02%, in case of S. seenghala, tested variables divided into nine factors with two main components, accounted for 99.91% and in case of W. attu divided into four factors with first two main factors which accounted for 99.996% of cumulative variability.

The correlation for different physicochemical parameters between different sampling sites showed that pH with water temperature (r = 0.107) and dissolved oxygen (r = 0.905) was positively non-significant while the correlation with electrical conductivity (r = -0.798), salinity (r = -0.888), total dissolved solids (r = -0.857), total alkalinity (r = -0.736) and total hardness (r = -0.499) was negatively non-significant. The electrical conductivity was positively correlated with all the physic-chemical parameters as with water temperature (r = 0.482), salinity (r = 0.925), total dissolved solids (r = 0.889), total alkalinity (r = 0.452) and total hardness (r = 0.906) and this correlation was non- significant. The correlation between the total alkalinity and total hardness was also positive and non-significant (r = 0.048).

The number of bands for PCR products in Channa punctatus ranged as low as three to a maximum of seven, with an average of 6 bands per primer. The number of polymorphic bands per primer was 1 to 3 with 14.29% to 50%. The number of bands in Channa marulius ranged as low as three to a maximum of seven, with an average of 6 bands per primer with 1 to 3 polymorphic bands. The polymorphic bands ranged from 14.29% to 50%. The number of bands in Rita rita ranged as low as three to a maximum of seven, with an average of 6 bands per primer with 1 to 3 polymorphic bands. The polymorphic bands ranged from 14.29% to 50%. The bands were as low as three to a maximum of seven, with an average of 6 bands per primer in Sperata seenghala with 1 to 4 polymorphic bands. The polymorphic bands in these populations ranged from 14.29% to 57.14%. The number of bands produced in Wallago attu, ranged from three to seven, with an average of 6 bands and 1 to 4 polymorphic bands per primer were observed. The polymorphic bands in these populations ranged from 14.29% to 66.67%.

The dendrogram for C. punctatus divided the randomly selected individuals of the five populations into four classes/clusters; 17 members in first cluster/class, 4

xv samples in second cluster/class and 2 samples in the third and fourth class/cluster. The randomly selected individuals of the five populations of C. marulius were divided into four classes/clusters; 21 in 1st, 2 in 3rd and one in 2nd and 4th class, each. The populations of R. rita were divided into four classes/clusters as in C. marulius. In S. seenghala four clusters with 21 in 1st, 2 in 2nd and one in 3rd and 4th class, each. In W. attu, into six classes/clusters. i.e., 16 individuals in 1st class/cluster, 3 individuals in 2nd and 4th Class/cluster, each, while in 3rd, 5th and 6th were only one individual, each.

The PCA regarding variability for C. punctatus indicated eleven factors, first four main factors accounted for 55.849%. The first and second group (F1 and F2) accounted for 14.881% each. Contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into four major variable groups, one group towards the positive side, one towards negative and two groups at the neutral distinction. According to the Kaiser (1958) criterion based upon the Eigen values, first three main factors out of eleven, accounted for 42.77% of cumulative variability in C. marulius. Contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into five major variable groups, two groups towards the positive side, two towards negative and one group at the neutral distinction. Rita rita populations were divided into nine factors/groups and first two main factors accounted for 30.968% of cumulative variability. Th trend divided the role of primers into five major variable groups, three groups towards the positive and one group towards the negative while the remaining at the neutral distinction. In same way, S. seenghala was devided into twelve out of which four main components all together accounted for 55.41% of the cumulative variation. The first and second group (F1 and F2) accounted for 15.717% and 14.044% respectively, of the cumulative variability. The trend for polymorphism divided the role of primers into seven major variable groups, two groups towards the positive side and four groups towards the negative side and one got the neutral distinction. Data in W. attu was divided into thirteen components and four main factors accounted for 50.37% of the cumulative variation in W attu. The trend divided the role of primers into four major variable groups, one group towards positive side, one groups toward negative side and two groups has the neutral distinction. The specimen Tsa1 shows complete distinction in all species except for C. punctatus.

xvi

Knowledge of genetic structure of the major River populations is helpful for management of the populations in order to maintain their genetic quality. In this study the results indicate good correspondence in the data analyses of morphometric parameters, and RAPD molecular markers using various statistical techniques with the exception of the distinction of individuals from different sites, which clearly indicated some environmental impacts, are likely influencing the genetic makeup within and between the local populations. This study also has provided the genetic information of the present fish populations from and how evolutionary processes are affecting the fish fauna. So this study along with the strengthening of the academic research area will also prove an applied research which will help the breeders to choose most fit candidates for the breeding program in the Pakistan.

xvii

Chapter 1

INTRODUCTION

Fish is an excellent source of high quality protein with high digestible contents and nutritional value. It contains many kinds of vitamins, particularly vitamin A and D and minerals like Phosphorus, Magnesium, Selenium, and Iodine. During the last few decades, research has revealed that fish meat is also an excellent source of omega3 fatty acids which are helpful in brain development and anticancerous in nature (Li and HU, 2009; KAWARAZUKA, 2010). In the year 2007, fish provided 15.7% of the global population’s animal protein, more than 1.5 million people with almost 20% of average per capita animal protein intake and 3 billion people with at least 15% of such protein. The high demands of fish led to the development of aquaculture which continued to grow with average annual growth rate of 6.6 percent, outpacing the human population growth rate (FAO, 2010). Same trend was observed in Pakistan where aquaculture had gained much impetus during the last few decades with the introduction of modern technology for seed production of major and Chinese carps in early seventies. Faster growth in the sector was the result of improvement in induced breeding technology and the establishment of many carp hatcheries throughout the country especially in the Punjab province (BHATTI et al., 1996). It has been estimated that 20% of the total inland fish production in the country is being contributed by the aquaculture (GOP, 2012). Aquaculture is relatively more developed in the provinces of Punjab and Sindh and to a lesser extent in Khyber Pakhtoonkhawah (AKHTAR, 2001), where freshwater fish culture is being practiced over an area of 40,000 acres producing more than 136,340 million tons of fish (GOP, 2012). The huge investments in farming have taken the shape of an industry with the ever increasing demand for fish.

In subcontinent, despite of large scale improvements in the field of aquaculture, the dominant culture system is the traditional polyculture of Indian major carps i.e. Labeo rohita, Cirrhinus mirgala and Catla catla. Exotic Chinese carps like 118

Cyprinus carpio, Ctenopharyngodon idella, Hypophthalmichthys molitrix and Aristicthys nobilis, have also been introduced in the culture system (BASAVARAGA et al., 1999). These were introduced in an attempt to diversify the existing culture system for increasing the yield. In spite of demand for carps, large segments of the population have preference for carnivorous fishes because of having lesser number of intramuscular bones, less fat and high protein contents as compared with carps (LING, 1977). Additionally, carnivorous fishes emerged as potential candidate for integration into existing polyculture system owing to its higher growth rates (ZHANG AND REDDY, 1991). The demand for these fishes increased substantially over the period of time and these species are fetching higher prices in national and international markets. In an attempt to cater these demands and to supplement the need for high value fish, many progressive aqua culturists introduced the local carnivorous fishes (Channa marulius, Sperata seenghala, and Wallago attu) in the existing polyculture systems and have reported attractive profits. The potential for culture of carnivorous fishes has not been fully exploited despite of their preferential characteristics (SAMANTARAY AND MOHANTY, 1997; ALI, 1999).

In Pakistan, fish fauna of River Indus is poor as compared to other rivers of the Asia viz. Brahmaputra, Ganges, Mekong, Salween, Hwang Ho and Yangtze, which with exception of Ganges originate from same geographical location of Tibetan highland Plateau. The length, drainage area, mean water discharge, slope, water temperature and sediment load of each river is variable which directly influences the diversity of Riverine ecosystem (WELCOME, 1985). The carnivorous fishes namely Channa marulius and Channa punctatus (Family Channidae), Wallago attu (Family Siluridae), Sperata seenghala and Rita rita (Family Bagridae) are important species of rivers in Indus basin and connected water bodies in Pakistan (MIRZA, 2003). These fish species also occur widely throughout the Indian subcontinent including India, Bangladesh and Nepal.

The Family Channidae consists of two genera viz., Channa and Parachanna (NELSON, 1984). The Asian genus Channa, which presently contains 26 valid species, is widely distributed in Iran, southern Asia and the Far East

119

(MUSIKASINTHORN, 2000; BERRA, 2001; MUSIKASINTHORN AND TAKI, 2001; COURTENAY AND WILLIAMS, 2004). The African genus Parachanna contains three valid species, which are restricted to central West Africa (BONOU AND TEUGELS, 1985; TEUGELS, 1992). The fishes of genus Channa, C. marulius (HAMILTON, 1822) and C. punctatus (BLOCH, 1793) (snakehead, murrel) are prominent tropical freshwater fishes widely used for medicinal and pharmaceutical purposes (MAT JAIS et al., 1994; MICHELLE et al., 2004), esteemed table fish and also used for ornamental trade (NG AND LIM, 1990; COURTENAY AND WILLIAMS, 2004).

Members of this carnivorous air-breathing genus are commonly found in a range of water bodies which include rivers, swamps, ponds, canals, drains, reservoirs, rice fields etc. ranging from southern Asia, southern China, Indochina to the Sunda Islands (MOHSIN AND AMBAK, 1983; LEE AND NG, 1994; HOSSAIN et al., 2008). In, Pakistan these fishes are common in natural water bodies and are normally marketed alive from the catch (MIRZA, 1982). Their contribution in aquaculture itself is only significant in certain neighboring countries such as Thailand, Taiwan, the Philippines, Vietnam, Malaysia, Cambodia and India (WEE AND TACON, 1982; HOSSAIN et al., 2008; JAMALUDINE et al., 2011).

Murrels are characterized by peculiar morphological features, such as elongated cylindrical body, long and entirely soft- rayed dorsal and anal fins, some Asian species entirely lack pelvic fins (ZHANG et al., 2002; MUSIKASINTHORN, 2003)., a large mouth with well-developed teeth on both upper and lower jaws, and an accessory air-breathing apparatus known as the suprabranchial organ (MUSIKASINTHORN, 1998; 2003). They have flattened heads; possess large scales on their heads and eyes being located in the dorsoventral position on the anterior part of the head. C. marulius is the fastest growing species among murrels reaching a length of 120–122 cm (BARDACH et al., 1972; TALWAR AND JHINGRAN, 1992) and have even been reported at an altitude of 475 m above mean sea level (MUNRO, 1955). It is considered to be a local migrant, and travels for a short distance for feeding purpose or for locating suitable breeding grounds in new water

120

bodies to avoid stress conditions of existing habitat. In polyculture system, this fish is also used as police fish to control the number of young tilapia in a fish farm (CURZ AND LAUDENICA, 1980).

Bagridae family, comprising of 27 genera (six in Asiatic region) is widely distributed in Asia and Africa (TALWAR AND JHINGRAN, 1992; MIRZA AND BHATTI, 1995). In subcontinent, catfishes belonging to thirteen families have been documented. Most of them are confined to fresh water but some are also found in marine water. These catfishes are among the well marketable freshwater fishes in south-east Asia. Riverine fisheries in the region support the fisheries of important Bagrid catfishes mainly Mystus seenghala, Mystus aor and Rita rita (MIRZA AND BHATTI, 1993; YADEV, 2006).

The giant river catfish, Sperata seenghala (SYKES, 1839) is easily recognized by its broad, spatulate snout with smooth upper surface, brownish-gray back, silvery flanks and belly and a dark well-defined spot on the adipose dorsal fin (FROESE AND PAULY, 2012). It is an important commercial species, contributing substantially to the total inland capture fish production in South Asia. According to MOHANTY et al. (2012), Sperata seenghala is a good source of lean meat and trace elements, especially Zinc and Iron. TRIPATHI (1996) considered M. seenghala and M. aor among the potential species of catfishes for future aquaculture plans.

Rita rita (HAMILTON, 1822) has elongated body with depressed head and strong dorsal spine. Pectoral spine is shorter than dorsal spine. Its mouth is transverse with 3 pairs of barbles and nostrils are wide apart. Lateral line is straight ending near forked caudal fin. Body color is silvery green or greenish above and flanks, sometimes brownish/blackish. The body colour is dull white from below.

Wallago attu (BLOCH & SCHNEIDER, 1801) is a genus of catfishes (order Siluriformes) of the family Siluridae, or "sheatfish great withe". The fish is commonly known by its genus name, Wallago. Though the genus contains more than one species, name "wallago" is also used as a common name for Wallago attu. Found in large rivers and lakes, it can 121

reach 2.4 m (8 feet) total length. This south Asian fish is found in Nepal, Bangladesh, Myanmar, India, Pakistan, Thailand, Vietnam, Cambodia, the Malay Peninsula and Indonesia. It is also reported from Laos and Afghanistan.

Wallago attu is demersal, freshwater, brackish fast running as well as sluggish fish of deep and shallow pools, beels, haors, rivers and streams. It is voracious, carnivorous and predatory fish. It is also called fresh water shark on account of its vicious biting and feeding habits. Elongated body is laterally compressed. Eyes are small. Mouth wide, its gape extends posteriorly beyond the eyes. Barbels are two pairs; among them, maxillary pair is long and extend posteriorly to well beyond origin of anal fin and the mandibular pair is much shorter, about as long as snout. Dorsal fin is short. Pectoral spine is weak. Caudal fin is deeply forked. Body color greyish or yellowish grey in above and whitish in below but the fins are grey (Rahman, 2005 and Talwar and Jhingran, 1992). Wallago attu has undergone significant decline due to overexploitation as a food fish throughout its range and thus has been included in IUCN red list of threatened species (Ng, 2010)

Biodiversity is essential for stabilization of aquatic ecosystem and protection of overall environmental quality (EHRLICH AND WILSON, 1991). It has been shown that habitats with greater biodiversity are more resilient, that is, they are better able to adjust and recover from various disturbances because of different species may perform overlapping functions in a biologically diverse and complex ecosystem, a disturbance that affects one species may have lesser impact on the ecosystem as a whole (SHINDE et al., 2009).

The genetic variability within population is extremely useful to gather the information on individual identity, breeding patterns, degree of relatedness and genetic variation among them (SCHIERWATER et al., 1994). Genetic diversity is an important characteristic for the short term fitness of individuals as well as for long term survival of the population, permitting adaptation to the changing environmental conditions. The study of genetic variability is, therefore, of prime importance for genetic approaches to fish conservation or breeding, which depend on knowledge of the amount of

122

variation existing in a local reproductive unit (CARVALHO, 1993). This type of information is useful for the development of management strategies that will conserve the biodiversity associated with different species, sub-species, stocks and races (TURAN et al., 2005).

Morphometric analyses is very useful for separating species, populations and races in the past and have been widely used for the identification of different fish stocks (TURAN et al., 2004, 2005). Morphometric variations between stocks provide a basis for stock structure and might be useful for studying short term, environmentally induced variation, for example, in fisheries management (BEGG et al., 1999). Morphometric studies of fish populations are also very important for understanding the interactive effect of environment, selection and heredity on the body shapes and sizes within a species (CADRIN, 2000). Information on the morphometric measurements of fishes and the study of statistical relationship among them are essential for taxonomic work (NAREJO et al., 2008). Several studies on the comparative morphometric of different fish populations have been conducted (NAKAMURA 2003; TURAN et al., 2005; IBANEZ- AGUIRRE et al., 2006).

A number of researchers are of the opinion that unplanned and unmanaged usage of a fish species in hatchery programs can also lead to inbreeding depression, with a potential reduction in fecundity, adaptation ability and survival rate (BEAUMONT AND HOARE, 2003; SUN et al., 2004). Keeping in view these problems, it becomes pertinent to conserve and manage the fish stocks effectively by using the vital information about relevant population genetics specifically through assessment of their genetic structure and diversity for potential brood-stock identification. Previous studies typically had a focus on reproductive biology (ALI, 1999), breeding (HANIFFA et al., 2000), medical and pharmaceutical properties (BAIE AND SHEIKH, 2000; MICHELLE et al., 2004), biochemical composition (ZURAINI et al., 2006; ZAKARIA et al., 2007), ecology (LEE AND NG, 1994; AMILHAT AND LORENZEN, 2005), diet (ROSHADA, 1994; ARUL, 2008) and morphological characters (CHANDRA AND BANERJEE, 2004).

123

Two different approaches are in practice for studying the genetic variability in aquatic animals. The first approach is usage of morphological characters to ascertain the genetic difference among populations. This approach used the assumption that animals with the same morphometric characteristics constitute one stock. This is relatively easy approach and is used widely in fishery stock differentiation studies (AVSAR, 1994; RASOOL et al., 2012a, 2012b and 2012c). The second approach is more sophisticated and used genetic markers for assessing the genetic variability among populations and stocks. AGNESE et al., (1997); DELLING et al., (2000); POULET et al., (2004) advocated that the use of molecular markers would be a very useful method to confirm the observed phenotypic differences between different geographical regions, and to facilitate the development of management strategies and future use of these species for intensive aquaculture plans

In molecular genetic characterization, random amplified polymorphic DNA (RAPD) and inter-simple sequence repeat (ISSR) are being widely used in studies to characterize genetic divergence within and among the populations or species of various animals. Genetic variations can be detected in different populations using various DNA markers like restriction fragment length polymorphism (RFLP) (POGSON et al., 1995 AND OKAZAKI et al., 1999), microsatellites (APPLEYARD et al., 2001; BEACHAM et al., 2002) and randomly amplified polymorphic DNAs (RAPD) (WILLIAMS et al., 1990; LEHMANN et al., 2000, KHALEDI et al., 2012 and RASOOL et al., 2012a, 2012b and 2012c).

RAPD technique is one of the most frequently used molecular methods for taxonomic and systematic analyses of various organisms (BARTISH et al., 2000), and provides important applications in fish species (DINESH et al., 1993, 1996; JOHNSON et al., 1994; CACCONE et al., 1997; BAGLEY et al., 2001; BARMAN et al., 2002; ALMEIDA et al., 2003; SANDOVAL-CASTELLANOS et al., 2007). The ease, simplicity and low cost involved in this technique make it ideal as number of polymorphic markers can be produced easily with no prior knowledge about the genetics of the organism.

124

In Indus river and its tributaries the significant work on the ichthyodiversity has been undertaken by several researchers (AHMAD, 1943, 1960, 1963; BANARESCU AND NALBANT, 1966; BUTT AND BUTT, 1988; HUSSAIN, 1973; QURESHI, 1965; SALAM et al., 1997; UROOJ et al., 2011). Most of this work was related with the ichthyodiversity and faunal description of the water bodies. Very little work was done on the differentiation of stocks and populations based on either morphometric or molecular approaches in Pakistan. With the decreasing flows of Indus River system and ecological changes due to extended dry periods combined with overexploitation have led to decrease in fisheries resources. There is an urgent need to understand the changes in genetic diversity of the fish populations and stocks of this river system. The deficient literature in this field made it expedient to study the genetic diversity of the fish stocks/ populations in the major rivers of the country. Keeping in view the findings of previous studies and prevailing situations the present study was planned in order to test the hypothesis that the representatives of the families, Channidae (Channa punctatus; Channa marulius), Bagridae (Sperata seenghala, Rita rita) and Siluridae (Walla attu) are gaining some distinction due to some environmental changes and other unknown factors in natural reservoirs. The specific objectives were:

1. To evaluate the use of the RAPD assay to generate the species specific profile for five carnivorous fishes named Channa punctatus; Channa marulius, Sperata seenghala, Rita rita and Walla attu collected from five sites from River Ravi, Jhelum, Chenab and Indus of Punjab-Pakistan.

2. To estimate the genetic similarity/variation among the five carnivorous fishes i.e., C. punctatus; C. marulius, S. seenghala, R. rita and W. attu from different water bodies of Punjab-Pakistan.

3. To examine the genomic variation among these fishes based on RAPD data and the population genetic structure for conservation of these fish species

125

CHAPTER 2

REVIEW OF LITERATURE

The literature surveyed for the present study “Genetic variability in some commercially important carnivorous fishes by molecular markers in Punjab, Pakistan” is described under the following headings

2.1 Brief History 2.2 Phylogenetic Diversity 2.3 Use of RAPD Markers 2.4 Other Methods of Genetic Diversity Studies

2.1 Brief History

The application of molecular genetics in fish studies is relatively of recent origin. But yet there is not attempt to characterize genetically any of the fish species in Pakistan. No reports are available on whether there exist genetically different populations among the members of a given species of carnivorous identified earlier in any of the major River systems. Recently, RASOOL et al., (2012a, 2012b, 2012c) has performed different studies on two major carps i.e., Labeo rohita and Cirhina mirgala but the genetic studies regarding the carnivorous fish species is still lacking in the country. The main reasons for this seem to be prioritization of problem/tasks in the fisheries sector. The priorities in fisheries, not only in subcontinent but also in other developing countries in Asia were different, ever since systematic research has been initiated. The need was to identify the economically important species of fish and study their biology and methods of propagation under controlled conditions. Once this was established, the fisheries workers were busy in developing, technologies for carp/fish farming systems to boost the production. Except for the work on hybridization, no importance

126

was given to any aspect of genetics what so ever, during previous years. The composite or multi-species culture technologies so far developed are based on species manipulation and application of certain management practices. These technologies, no doubt have boosted the fish culture several folds. However, at present it is felt that any further improvement in fish production may not be possible with these technologies and the importance of other aspects such as genetic quality and improvement of the candidate species by fully exploiting their hitherto untapped genetic potentials.

The urgency of acquiring knowledge about the existence of different fish stocks/populations among different species of economically most important fish has been realized. Lack of appropriate and accurate methodologies for genetic characterization and identification at population/species level has also been one of the main constraints to fully utilize the genetic resources (OKUMUS AND ÇIFTCI, 2003). The various methods available earlier, much before the advent of biochemical and molecular (DNA) techniques for stock identification or to study the existence of different populations in a given species, were only the morphometric measurements and meristic counts. These methods however, do not provide the degree of polymorphism distinguishable by modern methods especially within species (FAO, 2010).

The progress in the field of molecular genetics by the discovery of molecular DNA markers for the detection of polymorphism in the total genomic DNA or mtDNA has made it easier for the correct identification of fish populations. The use of DNA technology is gaining momentum since 1990s, many laboratories and institutions have come up in fisheries and are involved in research in this direction (PADHI AND MANDAL, 1995a). The advances in this field are by randomly amplified polymorphic DNA (RAPD) analysis by use of RAPD markers and restriction fragment length polymorphism (RFLP), analysis of genomic DNA or mtDNA by the prepared random oligonucleotide primers for the polymerase chain reactions (PCR) with the gel electrophoresis (PRENTIS AND MATHER, 2008). The RAPD technique is mostly used to identify the genetic populations of the quantitative traits loci (QTL).

127

LIU et al., (1994) states that six to seven primers are required to estimate the genetic variability within and between populations of a highly polymorphic species. They reported that the molecular evidences need to be collected for the differentiation in the morphological characteristics of the populations of same species.

2.2 Phylogenetic Diversity

During the last few decades, there is progress in the field of genetic diversity studies through the advancement in the fields of cytogenetic, biological, chemical and many molecular study methods. These are very good methods but the only utility of isozymes in the fishes, very common during this period, the prototype of it cannot identify the changes on the specific locations. In the same way the diversity studies via conservative amino acids replacement and their detection in the fishes for genetic studies are not enough to expose the differences in the genetic makeup which are actually represented in the morphological characteristics differentiation in the individuals of the same species (PADHI AND MANDAL, 1995b).

FERGUSON et al., (1995) reviewed the use of genetic markers to study the conservation of fish populations. They inferred that the genetic study techniques could be used in research on brown trout and Atlantic salmon to compare the resolution and applicability of allozyme, mitochondrial DNA and minisatellite (variable number of tandem repeats) markers for studies on population structuring, genetic variation within populations, and the impact of the accidental and deliberate introduction of non-native salmonids on the genetic make-up of natural populations.

CADRIN (2000) concluded from his experiment that it is often difficult to explain the causes of morphological differences between populations. These differences may be genetic differences, or they may be associated with phenotypic plasticity

128

in response to different environmental factors in each area (MURTA, 2000). BHARDWAJ (2005) classified as grossly polluted river Sutlej and Beas as a relatively clean river. It could be the addition of industrial wastes in large quantities from the sides of the river Sutlej responsible for the increased alkalinity and BOD at Harike site (DHILLON AND KAUR, 1996). SIN (1997) reported that phenotypic changes, such as increased growth rate, are usually more pronounced in transgenic fish than those obtained by artificial selection or efficient feeding.

VIGNAL et al., (2002) published a comprehensive review of use of a bi-allelic type of marker called Single Nucleotide Polymorphism (SNP) in animal research. According to these researchers the technique can provide valuable data on associations between specific genes or other DNA structures and phenotypes of population and genome dynamics. Subsequently, ALI et al., (2004) published a review of RAPD technique and inferred that it make ideal for genetic mapping, plant and animal breeding programs, and DNA fingerprinting, with particular utility in the field of population genetics.

CALLEJAS AND OCHANDO (2002) declared that the studies on phylogenetic relationships can be useful for planning future strategies for protection, because understandings inter- and intra-specific distribution of genetic variation is important for the development of conservation programs. TAINPEI (1991) observed that M. seenghala and M. aor are morphologically closer unlike all other types of Mystus sp. Also JAYARAM (1971), argued that morphologically M. seenghala and M. aor unique among other mystus species. Similarly, JAYARAM (1971) described the characteristics of osteological Wolf fishes subgenus Aorichthys mystus and promoted them to the rank of genus. The inclusion of these two catfish M. seenghala and M. aor in a separate genus Aorichthys (later on renamed as Sperata).

TURAN et al., (2005) found by experiments that morphometric comparisons of African catfish, Clarias gariepinus in different river systems in Turkey, showed significant deviation. Likewise, both morphological and genetic methods have been used to characterize different populations of Clarias gariepinus and Clarias anguillaris by AGNESE et al., (1997).

129

POULET et al., (2004) confirms association between genetic variation and morphological variation in natural populations. And both of these variations are used to make estimates of population differentiation was also suggested by BUTH AND CRABTREE (1982); IBANEZ et al., (2006).

VISHWANATH AND GEETHAKUMARI (2009) supports the morphological and osteological comparisons of previous authors in eight species of Channa representatives indicating two phylogenetic groups, gachua and marulius group. Gachua group has Channa amphibeus, Channa aurantimaculata, Channa barca, Channa bleher, Channa gachua, Channa punctatus and Channa stewartii, while other group, the marulius group includes Channa marulius and Channa striatus.

PERVAIZ et al., (2012) studied the meristic and morphometric characters of Indus Mahseer Tor macrolepis. They studied the fish during 2008 to 2009 from various sites of Attock district and adjoining areas by dividing the river into four sampling zones, each of 10 km, and one sampling zone was selected at the Hassan Abdaal, Pakistan. They studied 118 specimens for more than forty important morphometric and meristic parameters. Samples ranged from 12.32- 15.86 in total length (TL), 11.05-14.21 in fork length (FL) and 9.68-12.4 in standard length (SL). In the fish, gill rakers counted as 2-3/11-13, rostral barbel length (RBL) found slightly shorter than maxillary barbel length (MBL), no distinct stripes or spots present on body. They observed high level of significant relationships with total length (TL) and head length (HL) when compared to all other morphometric parameters studied.

2.3 Use of RAPD Markers

YOON AND KIM (2001) examined the genetic similarity and diversity of catfish Silurus asotus populations from two areas in western Korea using randomly amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR). Out of 20 random primers tested, 5 produced 1344 RAPD bands ranging from 8×2 to 13×6 polymorphic bands per primer. The polymorphic

130

bands in these populations ranged from 56×4% to 59×6%. Polymorphic bands per lane within populations ranged from 4×9% to 5×3%. The similarity within the Kunsan population varied from 0×39 to 0×82 with a mean (± SD) of 0×56 ± 0×08. Band sharing values 0×59 ± 0×07 within the catfish population from Yesan. They were of the opinion that the genetic similarity in populations may be caused because individuals were from the same habitats or by inbreeding during several generations. However, in view of band sharing values, polymorphic bands and also the specific major bands, significant genetic differentiation between these populations was present.

BARMAN et al., (2003) suggested that the special characteristics of the RAPD method (random, more genome characterized loci, the dominant type of markers, and the possibility of migration of non-homologous bands), RAPD analysis are limited only. Despite of these limitations, the RAPD analysis is used efficiently in different fish species for the initial assessment of genetic variability. ALI et al., (2004) explained that the main advantages of RAPD markers, i.e., the opportunity to work with anonymous DNA and relatively low cost, and it's quick and easy to produce RAPD markers. LEESANGA et al., (2000) observed genetic differentiation between the different populations of yellow catfish (Mystus nemurus) from Thailand.

HUANG et al., (2005) used RAPD markers to assess hybridization of endemic catfishes Clarias fuscus, C. mossambicus, and C. batrachus. They found that the unique RAPD markers generated from 3 PCR primers showed the presence of alleles in the genomes of C. mossambicus and absent in the genome of C. fuscus. Thus hybrids of C. fuscus and C. mossambicus, therefore, could be distinguished by the use of these specific molecular markers.

SIRAJ et al., (2007) used RAPD markers to examine the genetic relationships among three populations of two different color-types (silver-bronze and reddish) of ikan kelah (Tor tambroides). They sampled sixty three individuals of the kelah from Sia River of Pahang and Kampung Esok River of Negeri Sembilan (silver-bronze) and Nenggiri River of Kelantan (reddish). They used twelve RAPD primers which generated a total of 226 scorable loci with 100% polymorphism across

131

the sixty-three individuals. The RAPD banding patterns and sizes ranged from 4 to 17 and from 100bp to 1500bp, respectively. The intra-population UPGMA dendrogram produced two major clusters, with the Nenggiri River (Kelantan) samples formed a sub cluster in both major clusters dominated by the Pahang samples (Cluster 1) and N. Sembilan samples (Cluster 2), respectively. They found that the Kelantan samples were genetically closer to the N. Sembilan samples than to the Pahang samples in the inter-population UPGMA.

It was suggested by PHALE et al., (2009) that the species-specific markers, interspecific hybrid gene and flow identification can be used for the genetic variation studies while studying the Indian major carps. NAGARAJAN et al., (2006) studied genetic variations between Channa punctatus populations collected from three rivers of south India using randomly amplified polymorphic DNA (RAPD) in sixty samples from each population. They used six primers for genetic variation studies. The amplified RAPD bands were 42 in C. punctatus. The total number of bands observed ranged from 34 in the Quilon population to 37 in the Thirunelveli population. Among the three populations, the highest genetic identity (0.9231) was found between Thirunelveli and Quilon populations. Similarly, four AFLP primer combinations and nine RAPD primers detected a total of 158 and 42 polymorphic markers, respectively. The results of AFLP and RAPD analysis provide similar conclusions as far as the population clustering analysis is concerned. Both marker systems revealed high genetic variability within two populations. Three subgroups each from the Kedah, Perak and Sarawak populations were detected by AFLP but not by RAPD. Unique AFLP fingerprints were also observed in some unusual genotypes sampled in Sarawak (CHONG, 2000).

MANDAL et al., (2009) detected the intra-specific variations in different populations of Chitala chitala by using different kinds of molecular markers and some enzyme systems. The results for variation were postulated by 27-enzyme systems, 6- random amplified polymorphic DNA RAPD-primers and 2-microsattelite loci by amplifying the DNA by PCR reactions.

132

The calculated FSt values for different subdivisions remained as, 0.1235 for RAPD and a collective FSt of 0.0344 for microsatellite loci. They were unable to get any polymorphism in the fishes from thirty eight allozyme loci.

GARG et al., (2010) used RAPD-PCR to analyze the genetic variation of two populations of Mystus vittatus (Bloch) collected from two different reservoirs and found that the primers produced 388 scorable DNA fragments of which 252 (64.98%) are polymorphic, 182 (46.9%) are monomorphic and 14 (3.61%) are unique. They found that RAPD banding patterns showed variations between and within populations while the morphological variations were negligible.

SULTANA et al., (2010) used RAPD fingerprint method to study the genetic variation in five different populations of Heteropneustes fossilis. They used four primers which produced 31 scorable bands and 26 (83.87%) were polymorphic in nature. The level of intra-population similarity indices in almost all the populations found to be similar. The dendrogram based on Nei’s original measure of genetic distance indicated the segregation of five populations into two distinct groups.

GARG et al., (2012) studied the genetic diversity among five geographically isolated populations of a freshwater cat fish, Sperata seenghala, using RAPD-PCR primers as genetic markers. 60 accessions were collected form five lentic water bodies of Madhya Pradesh namely Bhadbhada reservoir (n=08), Mohinisagar reservoir (n=11), Bansagar reservoir (n=15), Bargi reservoir (n=11) and Gandhisagar reservoir (n=15) constructed on Kolans, Sindh, Sone, Narmada and Chambal Rivers, respectively. Ten random primers were primarily scored in all 60 individuals of which five primers gave scorable results. By comparing the RAPD banding patterns, variations were found between and within the population. Bhadbada reservoir, Mohinisagar reservoir, Bansagar reservoir, Bargi reservoir and Gandhisagar reservoir showed genetic polymorphism as 61.09, 87.59, 93.96, 88.46 and 71.20% respectively, which clearly indicates that, Bansagar reservoir may be a good habitat as much as to conservation concern. The un-weighted pair group method with averages (UPGMA) dendrogram generating using Jaccard’s coefficient showing all five populations distributed in the five clusters indicating each cluster for each population who represented separate gene pool. Principal component analysis (PCA)

133

variance showed comparative statistics as component 1 and component 2 where Bahsagar population is totally isolated with rest of four populations. However, Non-Metric Multidimensional Scaling (NMDS) showed overlapping these populations with each other’s.

DANISH et al., (2012) perform the study to identify genetic relationship and diversity of Clarias batrachus populations collected from hatchery and wild stocks through RAPD. A total of 1376 RAPD bands ranging from 0.2 to 1.36 kb were amplified using five selected primers. The number of amplification products produced by a primer ranged from as low as three to a maximum of 18, with an average of 16 bands per primer. The polymorphic bands in these populations ranged from 56.4 to 59.6%. Polymorphic bands per lane within populations ranged from 4.88 to 5.3%. The similarity within the population from wild varied from 0.40 to 0.83 with a mean ± SE of 0.57 ± 0.08. The Jaccard’s similarity coefficient ranged from 0 to 0.27. At 0.06 similarity coefficient, two major clusters were formed, which indicates that the genotypes belonging to same clusters were genetically similar and those belonging to different clusters were dissimilar. Significant (P < 0.05) population differentiation indicated some degree of intra- and inters- population genetic variations in two populations of fish. This might be due to difference in habitat and breeding strategies between the two populations.

GIRI et al., (2012) studied the genetic relationship in six different populations of Acorus calamus through randomly amplified polymorphic DNA (RAPD) markers. A total of 574 DNA fragments ranging from 281 to 1353bp were amplified using 10 selected primers. The number of amplification products produced by a primer ranged from as low as six to a maximum of 13, with an average of nine bands per primer. The cluster analysis revealed three major clusters; the first cluster contained samples collected from Luckhnow and Paonta Uttaralkand, India. The second cluster had single sample from Bangalore, India and the third cluster contained sample collected from Solan, Nauni and Hissar, India. The similarity coefficient value ranged from 0.97 to 0.88. The highest similarity coefficient (0.97) detected between samples from Nauni and Hissar as well as between Solan and Hissar and the lowest (0.88) was detected between the pairs Luckhnow and

134

Bangalore, Luckhnow and Solan, Bangalore and Solan. The level of polymorphism in their study was very low which showed that samples used for analysis would have close relationship.

RASHID et al., (2012) conducted the RAPD fingerprinting analysis to provide baseline information on the population genetic structure in two wild (Chalan beel and Tola haor) and one from Brahmaputra Fish Seed Complex-BFSC hatchery populations of O. bimaculatus. Five selected decamer random primers amplified a total of 34 RAPD bands among which 24 were found to be polymorphic. The percentage of polymorphic loci, intra-population similarity indices, gene diversity and Shannon’s information index values were 64.71%, 77.57%, 0.249±0.216 and 0.365±0.303 for Chalan beel, 58.82%, 75.45%, 0.219±0.215 and 0.322±0.304 for Tola haor population and 52.94%, 86.49%, 0.214±0.219 and 0.311±0.312 for the hatchery, respectively. The coefficient of population differentiation (PhiPT) between the Chalan beel - BFSC and Tola haor - BFSC pairs were found to be significant. The gene flow (Nm) between the population pairs ranged from 1.899 to 5.052. The highest inter-population similarity was found between Chalan beel-BFSC populations. Among the three populations, the highest genetic distance (0.157) was found between Tola haor and the BFSC population. The results indicated a substantial level of genetic variation in the endangered O. bimaculatus populations in Bangladesh and significant differentiation among the populations

KUMLA et al., 2012 studied Random amplified polymorphic DNA (RAPD) and inter-simple sequence repeat (ISSR) markers to investigate the genetic structure of four subpopulations of Mystus nemurus in Thailand. They selected 7 RAPD and 7 ISSR primers and out of 83 total RAPD fragments, 80 (96.39%) were found polymorphic and of 81 total ISSR fragments, 75 (92.59%) were polymorphic loci. Genetic variation and genetic differentiation obtained from RAPD fragments or ISSR fragments showed similar results. Percentage of polymorphic loci (%P), observed number of alleles, effective number of alleles, Nei’s gene diversity (H) and Shannon’s information index revealed moderate to high level of genetic variations within each subpopulation and overall population. High levels of genetic differentiations were received

135

from pairwise unbiased genetic distance (D) and coefficient of differentiation. Mantel test between D or gene flow and geographical distance showed a low to moderate correlation. Analysis of molecular variance indicated that variations among subpopulations were higher than those within subpopulations. The UPGMA dendrograms, based on RAPD and ISSR, showing the genetic relationship among subpopulations were grouped into three clusters; Songkhla (SK) subpopulation was separated from the other subpopulations.

2.4 Other Methods of Genetic Diversity Studies

VRIJENHOEK reported in 1998 that the rainbow trout, O. mykiss, a migratory species, has 85% of its diversity within local populations, and15% between. The Yellowstone cutthroat trout, a non-migratory species, has 67·6% of its diversity within populations and 32·4% between populations. In contrast, the small topminnow, Poeciliopsis occidentalis has only 21·3% of its diversity within populations and 78% between populations.

WOLSTENHOLME (1992) found that mtDNA make considerable savings in terms of size and gene number, repetitive sequences would probably. MABUCHI et al., (2004) suggested that the tRNA pseudo genes in parrotfish Met. is necessary as the punctuation signal to neighboring ND2 gene transcription. This functional limitation for punctuation could also work for duplicated genes in tRNA channels for met. Random genetic alterations occur in the evolution of metazoan mtDNA, and some of them have the useful property that defines the monophyly of taxa that stocks them, which supports the molecular as synapomorphic character, can be seen strongly sister relationship between the two species.

KOCHER et al., (1993) reported that the Cytochrome b fragment amplified using universal primers is useful in detecting intraspecific differences in several ways. JOHNS AND AVISE (1998) used cytochrome b analysis to find out the genetic divergence in eighteen genera of fishes. CHONDAR (1999) showed that Murrel is a local migrant travel only a short

136

distance for feeding or for finding suitable breeding sites or in search of new water to avoid stress conditions in the existing ecosystem.

A sample of African Clarias catfishes from the Senegal River was studied using morphometric, allozyme variation, microsatellites and RFLPs of mitochondrial DNA. They all confirmed the presence of two species, C. gariepinus and C. anguillaris. The two species were closely related genetically and no diagnostic loci were found in allozymes and microsatellites studies. Two of the 11 haplotypes of mtDNA observed were shared by both species. Three of the four assays (morphometric, allozymes and microsatellites) allowed a precise characterization of both. One specimen occupied an intermediate position in the analysis of the data it was considered an F hybrid by the workers AGNESE et al., (1997).

RAO AND MAJUMDAR (1998). Studied tetra nucleotide microsatellite loci for Indian major carp Catla catla. The alleles were not prone to stutter artifacts, usually associated with dinucleotide loci, and are readily visualized with a fast silver- staining protocol after electrophoresis in non-denaturing polyacrylamide.

NAISH AND SKIBINSKI (1998) has reported the multivariate map representation of phylogenetic relationships application to tilapia fish. A map-like representation provided information additional to the tree representation. The current belief that Sarotherodon is closer to Oreochromis than to Tilapia is strengthened. But while it may be the link between these genera at the species level, it is not entirely distinct from Oreochromis at the molecular level. Further, Sarotherodon and Oreochromis species may have arisen from Tilapia in several speciation events. Some of the species interrelations agreed with inferences from morphological data, and disagreed with those from a consensus maximum parsimony (MP) tree. It is suggested that both, Chromido tilapia guntheri and Tylochromis jentinki are ancestral to different sub-groups of Tilapia, so that inferences from morphological studies and the consensus MP method are both partially correct. The graphical representation also suggests that the Nile tilapia strains in Asia may be derived from Egypt rather than from

137

Ghana. It is advantageous to use the map-like and tree representations together for maximum visual in formativeness and inference from the data.

GOPALAKRISHNAN et al., (2004) studied a total of eight polymorphic microsatellite loci obtained from genomic library of Indian feather back, Chitala chitala (order Osteoglossiformes, family Notopteridae) and analyzed 46 samples to determine genetic variation. They found a mean number of alleles per locus ranging from 4.50 to 5.25, and expected heterozygosity ranging from 0.124 to 0.852. Deviation from Hardy–Weinberg equilibrium expectations (P<0.002) was observed at loci sample 2, sample 9 from Bhagirathi and sample 9 of Brahmaputra. They found microsatellite loci promising for population genetics studies of C. chitala and related species Notopterus notopterus.

PUNIA et al., (2006) studied sixteen primer sequences, from three cyprinid fishes to amplify homologous microsatellite loci in Gonoproktopterus curmuca. Six primers provided successful amplification. Total five loci were polymorphic. The mean observed heterozygosity was 0.293 and 0.471. Significant genetic heterogeneity (P<0.05) between the two sample sets provides the evidence that the identified microsatellite markers were promising in analyzing intraspecific divergence in G. curmuca.

AMBAK, et al., (2006) studied the genetic relationships among four populations of Channa striata distributed in Malaysian Peninsula. They used a total of 8 primers produced 42 polymorphic bands among all populations. Using this technique, only slight differences in genetic diversity were detected among locations of Channa striata within Peninsular Malaysia although the diversity of all locations analyzed together differed from the reference populations. Differences in genetic distance between populations may be due to selection pressure of pollutants on fish.

In the study on genetic variability of four fish species (Pimelodus maculatus, Prochilodus lineatus, Salminus brasiliensis and Steindachneridion scripta) collected from Uruguay River basin RAMELLA in 2006, made analysis using the RAPD technique. He obtained a total of 118 amplified fragments, 11 for P. maculatus, 29 for P. lineatus, 45 for S. brasiliensis 138

and 33 for S. scripta. Any monomorphic profile was not found in the studied species, except for S. brasiliensis, which presented seven monomorphic bands for Saltinho population. They concluded that all species showed a high level of genetic variability among individuals.

Genetic variability in three clariid species, Clarias batrachus, C. macrocephalus (Native species to Asia) and C. gariepinus (introduced for culture in Asia) was investigated by allozyme electrophoresis and mitochondrial DNA through RFLP markers by the workers MOHINDRA et al., (2007). They analyzed sixteen gene loci from 12 enzyme systems. Fixed allelic differences were evident between pairs of species at least at four loci. In mtDNA RFLP analysis, eight composite haplotypes were observed and each species was characterized by a set of haplotypes. The UPGMA phylogenetic tree revealed three distinct.

In the study upon spotted murrel (Channa punctatus) HANIFFA, et al., (2007) collected samples from three rivers of Tamil Nadu and Kerala and investigated allozyme variation of C. punctatus by polyacrylamide gel electrophoresis. They used eighteen enzymes, but found only 10 (EST, PGM, G3PDH, G6PDH, SOD, GPI, ODH, GDH, XDH, and CK) consistent phenotypic variations. They estimated allele frequencies at 18 polymorphic loci representing 10 enzymes out of them two rare alleles, EST-4*C and G6PDH-2*C, were noted in the Tamirabarani and Kallada populations which were absent in the Siruvani population. The allele frequencies of the Tamirabarani and Kallada populations were similar, except for a few loci. They reported a maximum genetic distance (0.026) and FST (0.203) between the geographically distant Siruvani and Kallada populations among the three populations.

NGUYEN (2008) conducted a study on the highly fragmented variety of Tor douronensis, Cyprinidae for the population structure analysis in Sarawak, Malaysia through microsatellite DNA markers. They concluded that the conservation and the management of the fish populations especially those which are threatened by the fragmentation of the habitat is necessary and urgent need to save their existence. The results of the study showed that seven autosomal microsatellite

139

loci data gave consistency among the different geographical and fragments sites. The results also pointed out that there are two well defined classes of T. douronensis in Sarawak, named as the north-eastern and the south-western classes. In addition, a further subdivision was observed in each of the clusters distributed between River systems. Low levels of gene flow were also observed and migrants between habitat fragments were identified, possibly resulting from human mediated translocations.

Liu, et al., (2008) performed Inter-simple sequence repeat (ISSR) analysis in order to evaluate the genetic diversity of wild and hatchery samples of half-smooth tongue sole Cynoglossuss emilaevis. They collected a group of 200 genotypes belonging to four wild samples, Laizhou (LZ), Weihai (WH), Qingdao (QD), Rizhao (RZ) and one hatchery sample, Mingbo (MB) screened using 15 different ISSR primers. A total of 137 loci produced in the five studied samples and found 41.80%, 45.26%, 44.27%, 42.86% and 41.59% of these loci polymorphic over all the genotypes tested in LZ, WH, QD, RZ and MB samples, respectively. The number of polymorphic loci detected by single primer combination ranged from 2 to 7. The average heterozygosity in their study was 0.0710, 0.0814, 0.0793, 0.0727 and 0.0696, in LZ, WH, QD, RZ and MB samples respectively. The results showed a higher genetic diversity in WH samples including total number of ISSR bands (P<0.05), total number of polymorphic bands (P<0.05), average heterozygosity (P<0.05) and total number of genotypes (P<0.05) than all the other samples. According to them the hatchery samples (MB) showed the lowest genetic viability among the five studied samples.

The comparative assessment of genetic diversity using allozymes, random amplified polymorphic DNA (RAPD), and microsatellite markers was conducted in endemic and endangered yellow catfish (Horabagrus brachysoma) sampled from three locations in Western Ghats river systems of India by MUNEER, et al., (2008). They reported that Microsatellites show more polymorphism, having 100% polymorphic loci, whereas allozymes show the least (56%). In RAPD, 60.5% of fragments were polymorphic. Observed heterozygosity and FST values were very high in microsatellites, compared with

140

the other markers. Microsatellite and RAPD markers reported a higher degree of genetic differentiation than allozymes among the populations depicted by pairwise F ST/G ST, AMOVA, Nei’s genetic distance, and UPGMA dendrogram. The three classes of markers demonstrated striking genetic differentiation between pairs of H. brachysoma populations. They inferred from their study that there is a need for fisheries management, conservation, and rehabilitation of this species.

There was a long debate over the taxonomic status of Channa orientalis, i.e. whether it is a specific type, or only those without abnormal pelvic fins and the same species C. gachua one sympatrically distributed species, morphologically very similar, but has pelvic fins (PETHIYAGODA, 1991).

LAKRA et al., (2008) demonstrated that the success of conservation programs for the protection and effective management policies at the level of genetic divergence within and between species and the development of strategies to obtain the natural genetic diversity. BOWEN et al., (1992) found that mtDNA for solving problems related to population groups in cases where morphological analysis were either inadequate or controversial, as there is comparatively low level of variability and a slow rate of evolution TINTI et al., (2002) find that in the case of Sardina pilchardus, mtDNA Cytochrome b sequence analysis gave no strong hydrographic or environmental factors acted as sub structuring force in the Adriatic population of fish, and provide a valuable demographic information to define the pattern of exploitation of these fish species from a fisheries man management perspective.

The genetic diversity of the Ptychidio jordani was studied by the use of 13-polymorphic microsatellite markers. They collected samples from the three locations viz. Guiping, Yunan and Liuzhou from the Pearl River basin. The section bulk ranged from 108-288 while the allele numbers were 2-19. They observed high level of genetic variation by comparing the information about the average polymorphism contents and the observed average heterozygosity. The expected heterozygosity was 0.44, 042 and 0.50, respectively. The variation between the sub populations was low as indicated by the genetic variation coefficient values which were 0.0074-0.0156. Analysis of Molecular Variance showed also the low

141

variability with 0.53-percent diversity between sub populations and difference within the sub populations with value 99.47% and the FSt value 0.0053. They concluded that the reason for morphological diversity within the sub populations were as result from the variable environments. They found a significant correlation among the genotype of microsatellite loci H1j-038 and the ratio of stoutness and length of body. From these studies they also concluded that there is a link among the suggested loci and morphological characters (ZHU et al., 2009).

MUNEER, et al., (2009) used 7 RAPD and 7 ISSR primers to investigate the genetic structure of four subpopulations of Mystus nemurus in Thailand. They found, 80 (96.39%) polymorphic loci out of 83 total RAPD fragments, and 75 (92.59%) polymorphic loci out of 81 total ISSR fragments. Genetic variation and genetic differentiation obtained from RAPD fragments or ISSR fragments showed similar results. Percentage of polymorphic loci (%P), observed number of alleles, effective number of alleles, Nei’s gene diversity (H) and Shannon’s information index revealed moderate to high level of genetic variations within each M. nemurus subpopulation and overall population. High levels of genetic differentiations were received from pairwise unbiased genetic distance D in their studies and coefficient of differentiation. Mantel test between D or gene flow and geographical distance showed a low to moderate correlation. Variations among subpopulations were higher than those within subpopulations obtained from AMOVA. The UPGMA dendrograms, based on RAPD and ISSR, showed the genetic relationship among subpopulations and grouped the samples into three clusters; Songkhla (SK) subpopulation was separated from the other subpopulations. The candidate species-specific and subpopulation-specific RAPD fragments were sequenced and used to design sequence-characterized amplified region primers which distinguished M. nemurus from other species and divided SK subpopulation from the other subpopulations.

MUNEER, et al., (2011a) studied two species of yellow catfish, Horabagrus brachysoma and H. nigricollaris are categorized as ‘endangered’ and ‘critically endangered’ respectively in their wild habitat for genetic variation and phylogenetic relationships between these species of yellow catfish. They sampled the fishes from Chalakkudy River in the

142

hot spot of biodiversity-Western Ghats region, Kerala, India and got analyzed by using Random amplified polymorphic DNA (RAPD) and microsatellite markers. They found 85 RAPD and five microsatellites loci to analyze the genetic variation and phylogenetic relationships among these species. They reported that out of 85 RAPD loci produced only 52.94% were polymorphic whereas in microsatellite, all 5 loci were polymorphic (100%). Species-specific RAPD bands were found in both species studied. In microsatellite, the number of alleles across the five loci ranged from 1 to 8. The observed heterozygosities in H. brachysoma and H. nigricollaris were 0.463 and 0.443, respectively. Here, both RAPD and microsatellite methods reported a low degree of gene diversity and lack of genetic heterogeneity in both species of Horabagrus.

MUNEER, et al., (2011b) developed and used random amplified polymorphic DNA (RAPD) and microsatellite markers for the analysis of genetic variability in the critically endangered yellow catfish Horabagrus nigricollaris, sampled from the Chalakkudy River, Kerala, India. They detected eight RAPD and five microsatellite markers and in RAPD, found the 73 fragments 20.55% polymorphic, whereas 4 polymorphic loci (80%) in microsatellites. In microsatellites, the number of alleles across the 5 loci was 1-5, and the range of heterozygosity was 0.25-0.5. The mean observed number of alleles was 2.4, and the effective number was 1.775 per locus. The average heterozygosity across all investigated samples was 0.29, indicating a significant deficiency of heterozygotes in this species. They concluded that the RAPD and microsatellite methods report a low degree of gene diversity and lack of genetic heterogeneity in the population of H. nigricollaris which emphasize the need for fisheries management, conservation, and rehabilitation of this species.

In their study to estimate the population structure and phylogenetic relationships among samples of the Salmo trutta complex which inhabit the Balkan Peninsula APOSTOLIDIS, et al., (2011) used five random oligo decamers, 140 fish from seven populations and got 55 discernible DNA fragments, of which 50 (90.91%) were polymorphic. The statistical results indicated a low genetic diversity within populations with an average percentage of polymorphic bands (P) of 11.69% and a

143

Nei’s genetic diversity index (h) of 0.035, but at the same time high genetic differentiation among populations (F ST = 0.89). The distribution of genetic diversity among Balkan trout may result from their evolutionary history and reflects genetic drift coupled with bottleneck phenomena.

The study of HABIB et al., (2011) evaluated partial Cytochrome b gene sequence of mtDNA for determining the genetic variation in wild populations of C. marulius. Genomic DNA extracted from C. marulius samples (n = 23) belonging to 3 distant rivers; Mahanadi, Teesta and Yamuna was analyzed. Sequencing of 307 bp Cytochrome b mtDNA fragment revealed the presence of 5 haplotypes with haplotype diversity value of 0.763 and nucleotide diversity value of 0.0128. Single population specific haplotype was observed in Mahanadi and Yamuna samples and 3 haplotypes in Teesta samples.

SAINI et al., (2011) studied the phylogenetic relationships and genome specificity among six species of Bagrid catfishes (Mystus bleekeri, M. cavasius, M. vittatus, M. tengara, M. aor and M. seenghala) using RAPD markers as discriminating characters for the first time. 511 RAPD fragments were generated using ten decamer primers of arbitrary nucleotide sequences. Amplification reactions resulted in fragments ranging in length between 92 and 2,863 bp, which were assigned to 155 RAPD loci. Clearly resolved and repeatable bands were scored for their presence or absence in a binary matrix. Different RAPD profiles were observed for all the six Mystus species. In the study they generated three groups diagnostic, eleven group exclusive and 18 species-specific markers. UPGMA dendrogram constructed on the basis of genetic distance formed two distinct clusters, M. seenghala and M. aor form one separate cluster from other four species i.e., M. tengara, M. cavasius, M. bleekeri and M. vittatus. The inferences drawn from the above study clearly showed their genetic distinctness from the other four Mystus species and supported their inclusion into a separate genus, Sperata.

144

Molecular characterization and assessment of genetic variation in 28S region especially divergent domain D8 ribosomal DNA in fresh water catfish Ompok pabda, O. pabo, O. bimaculatus and W. attu carried by VERMA AND SERAJUDDIN (2012). The nucleotide sequences and GC% of divergent domain of 28S ranged from 302-473bp and 54% in all above subject species. It was found that the genetic distance between O. pabda and W. attu was highest, followed by O. bimaculatus and W. attu in the pair wise comparison. Since the genetic identity between O. pabda and O. pabo was highest, it can be said that these two species were most closely related as compared to the other species combinations. The sequence alignment suggested the presence of insertions/ deletions (indels) and substitutions (transitions and trans versions) in these regions. The indels were responsible for length variations. The study suggested that Ompok species has a closer relationship forming one cluster, in which O. pabo and O. pabda were closely related as compared to O. bimaculatus. The species W. attu radiated far and formed separate cluster was the conclusion of their study.

BHAT et al., (2012) studied the phylogenetic relationship among eight Channid species viz. Channa aurantimaculata, Channa bleheri, Channa diplogramma, Channa gachua, Channa marulius, Channa punctatus, Channa stewartii and Channa striatus using RAPD markers. They used eight random oligodecamers viz. OPAC03, OPAC05, OPAC07, OPAC09, OPAC19, OPA10, OPA11 and OPA16 to generate the RAPD profile. Estimates of Nei's (1978) unbiased genetic distance (D) demonstrated sufficient genetic divergence to discriminate the samples of different species and the values ranged from 0.3292 to 0.800 They strongly substantiate the view of earlier morphological and osteological studies of Channid species, the closer association among species in "gachua" and "marulius" groups

There is a report posted by Agricultural Science (2012) in which they reviewed briefly the fundamental principle of molecular systematics, research methods, phylogenetic analysis, molecular phylogeography, neutrality theory and molecular evolution as well as the minimum spanning network. Also they briefly reviewed the phylogenetic investigation progress of Pelteobagrus. Genus Pelteobagrus, belonging to Bagridae in Siluriformes, is endemic to East Asia. Among

145

them, five species distributed in China, are P. fulvidraco, P. intermedius, P. eupogon, P. vachelli and P. nitidus. They used mtDNA ND4 gene sequences to clarify the existing gaps in the phylogenetic relationships and genetic variability as well as population genetic structure of these species. Mitochondrial DNA ND4 (plus two 3′ flanking tRNA, Histidine tRNA, Serine tRNA) gene sequence variations of these five species from different drainages were studied in the present study. Wallago attu is designated as out-group in the study. About 702 base pair sequences were obtained through PCR amplification. Sequence variation analysis was conducted using MEGA 2.1 software. The parsimony informative sites were 144 bp and occupied 20.5%, and 206 bp, occupy 29.3% for the variable sites. The Kimura’s 2-Parameter sequence divergences were calculated and molecular phylogenetic trees were reconstructed by using the neighbor-joining (NJ) and maximum parsimony (MP) methods as well as Bayesian inference. The results show that in the three molecular cladogram, P. vachelli forms a sister group with P. nitidus; P. fulvidraco, P. intermedius and P. eupogon form a monophyletic group; (2) the phylogenetic position of P. fulvidraco, P. eupogon, and P. intermedius are not recognized. All the specimens were collected from the mainland China. Among them, 60 individuals of P. fulvidraco from 19 localities, 27 individuals of P. nitidus from 7 localities, 15 individuals of P. vachelli from 7 localities, and 13 individuals of P. eupogon from 2 localities. From the 60 individuals of P. fulvidraco, a total of 23 haplotypes are found and haplotype F2 was shared among almost all the localities. And 19 haplotypes for P. nitidus; 11 unique haplotypes in individuals of P. vachelli; and 7 haplotypes for the 13 individuals of P. eupogon, indicating a high degree of polymorphism of the four species. The statistic of molecular genetics and population genetics analyzed using ARLEQUIN 2.000. They found high haplotype and nucleotide diversity in the population of P. vachelli. In contrast, the higher haplotype diversity and the lower nucleotide diversity were in the population of P. fulvidraco, P. nitidus, and P. eupogon. P. fulvidraco had a unimodal mismatch distribution at the left side of the graph, and P. nitidus had a unimodal mismatch distribution, but both P. vachelli and P. eupogon had multimodal mismatch distribution. Based on the relationship t =2ut and the maturity, the expansion events most possibly occurred between 414,000 and 692,000 years ago in the P. fulvidraco population and 514,000 and 857,000 years ago in the P.

146

nitidus population. The results show that (1) P. vachelli showed a higher genetic diversity, whereas P. fulvidraco, P. nitidus and P. eupogon showed a lower genetic diversity; (2) minimum spanning network, the distribution of haplotypes and haplotype divergences represent a demographic expansion from a single ancestral lineage of P. fulvidraco between 414,000 and 692,000 years ago. For the statistics of other neutrality tests, Tajima’s D (-2.09, P=0.003), Fu’s Fs (-8, P=0) and SSD (0.0032, P=0.4310) were all significant for the P. fulvidraco; (3) there might be a population expansion of P. nitidus about 514,000–857,000 years ago. Fu’s Fs (-11.84, P=0.001) and SSD (0.0068, P=0.1870) are all significant for the P. nitidus. (4) In contrast, P. vachelli and P. eupogon may have experienced a long evolutionary history in a large stable population

RASOOL et al., (2012c) while working on clustering analysis for intraspecific variation studies amongst the populations of Cirrhinus mirgala reported that data of the morphometric parameters divided the populations of C. mrigala in to four major clusters or classes. They further reported that variance decomposition for the optimal classification values remained as, 27.28% for within class variation while 72.72% for the between class differences. The distance between the class/cluster centroids remained as; 50.820 for class one and two, 18.063 for class one and three, 14.564 for class one and four, 68.856 for class two and three, 36.708 for two and four while this distance between class three and four centroids was 32.408.

147

Chapter 3 MATERIALS AND METHODS

3.1 Study Sites

The samples of the target fish species were collected from Chashma Barrage at River Indus near District Mianwali, Qadirabad Barrage at Chenab River near District Gujranwala, Tehsil Wazirabad, Baloki barrage at Ravi River near Tehsil Bhai Pharo District Kasur, Trimu Barrage at the junction of Chenab and Jhelum Rivers near District Jhang and Taunsa Barrage at Indus River near Tehsil Kot Addu District Muzaffar Garh (Figure-1).

Baloki Barrage Chashma Barrage Qadir Abad Barrage Taunsa Barrage Trimu Barrage

148

Figure 1 Map showing the sampling sites at different Rivers in Punjab-Pakistan

3.2 Experimental Species

Following five carnivorous fishes Channa punctatus (Figure 2) and Channa marulius (Figure 3) belonging to the family Chanidae, Rita rita (Figure 4) and Sperata seenghala (Figure 5) belonging to family Bagridae and Wallago attu (Figure 6) of family Siluridae were selected for the present study.

Figure 2 Representative picture of the target species Channa punctatus (Daula)

149

Figure 3 Representative picture of the target species Channa marulius (Soal)

Figure 4 Representative picture of the target species Rita rita (Khagga)

150

Figure 5 Representative picture of Sperata seenghala (Singhara).

Figure 6 Representative picture of the target species Wallago attu (Mulley)

3.3 Sample Collection and Storage

The experimental species from the above mentioned sites were collected with the help of the local fishermen at the site and shifted to experimental laboratory by icing and were stored in the lab up to analysis at -80 oC.

151

3.4 Location of Research Facilities Used

The collected samples were studied at Molecular Genetics Lab of the Department of Wildlife and Fisheries, Faculty of Sciences and Technology, Govt. College University, Faisalabad.

3.5 Measurement of Physico-chemical Properties of Water

The physico-chemical factors/ water quality parameters are important for the population diversity of fish species in the Riverine ecosystems. In the study the procedure available in the manual "Standard Methods for the Examination of Water and Waste Water" by American Public Health Association (1975) was used.

The details of the procedures of quality parameters measured for the water samples collected from sampling sites following the A.P.H.A. (1975).

Collection of Water Samples

From each sampling sites along with the collection of fish samples for DNA extraction, water samples for limnological studies were taken in the 500 ml capacity polythene bottles. Some of the physiochemical characteristics i.e. water temperature, electrical conductivity, Dissolved Oxygen (DO), pH were recorded at the sampling sites. The samples were taken to the laboratory for analysis of total hardness, chlorides, magnesium, carbonates, bicarbonates, total alkalinity, total solids and total dissolved solids. The other relevant information about the collection of the water samples is given in the chapter 4.

152

A. Water Temperature The temperature of the water was noted which was indicated with the help of Dissolved Oxygen Meter (HI-9146) along with DO measurement by fixing the temperature factor at 0 oC.

B. pH To measure the pH (hydrogen ion concentration scale) the microprocessor pH meter (HANNA-HI-8520) was used after setting its range at “pH” point.

C. Dissolved Oxygen The dissolved oxygen was measured with the help of DO Meter (HI-9146) after setting its range “ppm” unit. The dissolved oxygen was measured directly at collection site by dipping the sensor of the meter into the water.

D. Electrical Conductivity (EC) EC is actually a measure of the ionic activity of a solution in term of its capacity to transmit current. Electrical conductivity was measured by conductivity meter, Condi 330i WTW 82362 Weilheim Germany in millisiemne.

E. Salinity Salinity is the saltiness or dissolved salt concentration of a body of water. It is a general term used to describe the levels of different salts such as sodium chloride, magnesium, calcium sulfates and bicarbonates. The salinity was measured by meter (Condi 330i, WTW 82362 Weilheim Germany)

F. Total Dissolved Solids (T.D.S.) Total dissolved solids were estimated by evaporation method. Water sample was first filtered and then 100 ml of this filtered sample was measured by graduated cylinder and was taken in the pre-weighed beaker and evaporated in the oven at 103 oC. After evaporation the beaker was again weighed and the T.D.S. was calculated by the following equation. 153

Total dissolves = Increase in weight of beaker in mg x 1000 solids (mgL-1) Volume of sample in ml G. Total Alkalinity Total alkalinity was measured by methyl orange indicator method (A.P.H.A. 1975). The sample for the alkalinity was analyzed soon after taking sample to the laboratory to avoid the denaturation.

Measurement of Total alkalinity was made by shifting 100 ml water sample measured by graduated cylinder into an Erlenmeyer's flask. The 4 to 8 drops of methyl orange indicator solution were added to the sample in the flask and sample was titrated against 0.02 N H2SO4 to end point (radish orange). The calculations for estimation of total alkalinity were made by using the following formula

Total alkalinity = Volume of H2SO4 (N of acid) (50) x 1000 (mgL-1) ) Volume of sample in ml

H. Total Hardness

The concentration of calcium plus magnesium is traditionally expressed as equivalent of CaCO3 was used as a measure of total hardness (A.P.H.A. 1975).

The pH of the water sample of 100 ml, measured by graduated cylinder into 250 ml Erlenmeyer flask, was raised by the addition of 2 ml of the buffer solution by mixing. After addition of buffer added 8 drops of Eriochrome Black T (EBT) indicator to it and was titrated with EDTA (0.010M) solution, till the solution was changed from wine-red to pure blue. The calculations for the measurement of total hardness were made by following equation. Total hardness = (Vol. of EDTA used) (M) (100) x (1000) -1 (mgL as CaCO3) Sample volume in ml 3.6 Morphometric Parameters of Experimental Species

154

The following morphometric parameters of the target species were measured and are presented in the chapter-4. A. Fish Wet Body Weight (g) B. Fork Length (cm) C. Total Length (cm) D Dorsal Fin Length (cm) E, Caudal Fin Length (cm) D. Anal Fin Length (cm) G. Average Pectoral Fins Length (cm) H. Average Pelvic Fins Length (cm) I. Head length (cm) J. Stoutness (cm) K. Adipose fin (cm) (Where available) 3.7 DNA Extraction Total genomic DNA isolation was carried out from the stored fish samples using procedure described by LOPERA- BARRERO et al., (2008). This procedure is based on the protocol given by (ALJANABI AND MARTINEZ, 1997) which was modified by the use of NaCl. A. Solution Preparation In this procedure, lysis buffer was used which carried 50 mM tris which was taken from a stock of 1 M at pH 8 tris buffer, 50 mM EDTA was taken from a stock of 0.5 M at pH 8, 100 mM NaCl taken from a stock of 5 M NaCl and 1% SDS. From this lysis buffer, working lysis buffer was prepared by adding 7 µl proteinase K from stock solution. Stock solution of the proteinase K was prepared by taking a buffer of 100 mM Tris-base, 50 mM EDTA, 500 mM NaCl and then 200 µgL-1 Proteinase K was added and dissolved.

155

B. Analytical Procedure About 1 g of the fish flesh was taken in a 1.5 ml eppendrof micro tube and homogenized in the 550 µL lysis buffer and then 7 µL of proteinase K buffer was added to the sample containing eppendrof micro tube. The contents of the tube were incubated in thermo-regulated water bath at 50 oC for 12 h. After this incubation 5M NaCl amounting 600 µL solution was added and mixed thoroughly and then centrifuged for 10 minutes at 12000 rpm.

A fresh eppendrof micro tube was taken and supernatant was transferred into it with the help of micropipette. Then the DNA was precipitated by addition of 700 µL absolute cold ethanol. After mixing the contents of the tube, it was incubated at -20 oC for 2 h. The tube was then centrifuged for 10 minutes at 12000 rpm to obtain the pellet of the DNA. All the liquid was discarded and 300 µl of 70% ethanol was added to remove salts. The washing with 70% ethanol was repeated and the pellet was dried by inverting the tube on a dry tissue paper. Air dried pellet of the DNA was dissolved in 80 µL TE buffer (10 mM tris and 1 mM EDTA). To remove the RNA from these preparations 1µl of 30 µgmL-1 of RNAs was added and incubated at 37 oC for one hour and then precipitated the DNA with 3.2 M sodium acetate and 2.5 ml volume of absolute alcohol. The pellet was centrifuged, washed with 70% ethanol, dried and dissolved in 50 µL sterilized TE buffer.

3.8 Quantification of DNA

Purity of DNA was checked for quantification by using UV spectrophotometer (U-2800, Hitachi) and agarose gel electrophoresis. For this purpose Optical Density (OD) value at 260 nm and 280 nm were taken and calculations were made to determine the concentrations of the DNA samples. For the assessment of the integrity of the DNA samples all the samples were sequestered on 1% agarose gel prepared in 0.5X TAE buffer which was obtained from 50X TAE stock solution prepared by dissolving 121 g tris base and 28.6 ml glacial acetic acid and 0.5 molar EDTA in water and raising its volume to 500 ml. The DNA samples were loaded into the gel after mixing with 10X DNA loading buffer with 0.21% bromophenol blue, 0.21% xylene cyanol FF, 0.2 molar EDTA and 50% glycerol.

156

3.9 Primer Selection

Twenty five decamer (Operon) primers designed by Gene Link Ltd., Hawthorne, were used initially in the study to amplify polymorphic DNA randomly. Ten of these with most scorable bands were used for further studies. The sequences of all the used primers are given below Table 1 Sequences of the used primers

Sr. No. PRIMER NAME SEQUENCE 1 OPB-01 GTTTCGCTCC 2 OPB-02 TGATCCCTGG 3 OPB-03 CATCCCCCTG 4 OPB-04 GGACTGGAGT 5 OPB-05 TGCGCCCTTC 6 OPB-06 TGCTCTGCCC 7 OPB-07 GGTGACGCAG 8 OPB-08 GTCCACACGG 9 OPB-09 TGGGGGACTC 10 OPB-10 CTGCTGGGAC 11 OPC-11 AAAGCTGCGG 12 OPC-12 TGTCATCCCC 13 OPC-13 AAGCCTCGTC 14 OPC-14 TGCGTGCTTG 15 OPC-15 GACGGATCAG

157

16 OPC-16 CACACTCCAG 17 OPC-17 TTCCCCCCAG 18 OPC-18 TGAGTGGGTG 19 OPC-19 GTTGCCAGCC 20 OPC-20 ACTTCGCCAC 21 OPD-01 ACCGCGAAGG 22 OPD-02 GGACCCAACC 23 OPD-03 GTCGCCGTCA 24 OPD-04 TCTGGTGAGG 25 OPD-05 TGAGCGGACA

3.10 PCR amplification of the Random Sequences from the fish samples

Polymerase chain reactions were devised with the help of the primers. Each reaction was performed in 0.2 ml PCR tube and 25 µL reaction mixtures. To prepare this 25 µL reaction, 2.5 µL 10x PCR buffer, 2 µL 1.6 mM MgCl2, 2 µL 10 nM primer, 2 µL 2.5 mM dNTPs, 0.3 µL 5 units/µL taq polymerase enzyme and 11.2 µL deionized double distilled water were mixed. In each reaction a negative control was also run using sterilized water as the template.

3.11 Profile of the PCR Reaction

PCR reaction was carried out in Personal Autorisieter Master Cycler of the EPPENDORF, Germany. Each reaction profile was of one cycle of 5 minute denaturation at 95 oC and then 35 cycles of 1 minute at 95 oC, 1 minute at 37 oC and 2

158

minutes at 72 oC and finally 20 minutes extension at 72 oC. Then the machine was allowed to hold the reaction contents at 22 oC for 30 minutes.

3.12 Analysis of the PCR products

All the PCR products were analyzed by sequestering them on agarose gel. For this purpose 1.5% agarose gel was prepared in TAE buffer as described in section 3.8. The DNA samples were then loaded on the gel using the DNA loading buffer. Each gel was run with 100 base pair DNA ladder in the left and right lanes or only on one side. These gels were visualized in UV light and photographs were taken by gel documentation system (WEALTEC, Dolphin-DOC).

3.13 Statistical analysis

Analysis of variance (ANOVA) for the different morphometric parameters and Pearson’s correlation among the physico- chemical parameters of water quality was done by Minitab statistical computer software. The XLSTAT 2012 version 1.02 of the computer software was used for the Pearson correlation analysis of the morphometric parameters of study. The same computer program was used for the difference in genotype occurrence on the basis of differences in morphometric parameters by Agglomerative Hierarchical Clustering (AHC) by following the Unweighted Pair Group Method with Arithmetic Mean (UPGMA). The Principle Component Analysis (PCA) was done on the basis of differentiation in morphometric parameters by Eigen values and differentiation into factors of the different genotypes from the different environmental conditions was done by correlation bi-plot/coefficient of the correlation (n) method in the same program. This software was also used to analyze the RAPD data for Jaccard’s coefficient by following the UPGMA for Hierarchical Clustering of the similar groups on the basis of similarity amongst the genotypes and the dendrogram generated which is presented in the next section. The PCA for grouping of the different genotypes from the different environmental conditions

159

was done by Spearman Varimax rotation method for bi-plot generation of the co-occurrence of the same genotypes with similar genetic properties and specificity of different primers in the same program.

160

Chapter 4 RESULTS

The data collected during the study was analyzed and is presented as follows

4.1 Morphometric parameters of experimental species

4.2 Principal Component Analysis for morphometric parameters of experimental species

4.3 Measurement of physicochemical parameters

4.4 DNA extraction

4.5 Quantification of DNA

4.6 Agglomerative Hierarchical Clustering (AHC) from PCR products/Agarose Gel Electrophoresis of experimental species

4.7 Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) data of experimental species

4.1 Analysis of Variance for Morphometric Parameters of the Experimental Species

The samples of the experimental species viz., C. punctatus, C. marulius, R. rita, S. seenghala and W. attu according to the plan of the study, 10-individuals of each species were collected from the study sites. The data of the morphogenetic parameters according to the physical appearance of the each species was recorded and the data is presented in appendix 1 to 5.

A. Morphometric Parameters of Channa punctatus

161

Body Weight

The average body weight of the Channa punctatus remained as; 135.5, 149.6, 120.8, 139.3 and 131.0 g, respectively, collected from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest weight containing individual fish was 180.0 g amongst samples collected from Chashma Barrage and Trimu Barrage and minimum weight individual was from Trimu Barrage with 90.0 g.

The data obtained for the body weight of Channa punctatus was subjected to Minitab computer software and the analysis of variance showed that there was non-significant difference (p > 0.05) in body weight among the sites.

Table 2 ANOVA for the body weight of Channa punctatus

SOV Df SS MS F-Value P-Value

Sites 4 4493 1123.130 2.534 0.053

Error 45 19949 443.302

Total 49 24441

SOV = Source of Variance, df = Degree of Freedom, SS = Sum of Squares, MS = Mean Sum of Squares and * = Significant

Total Length

The average total length of the Channa punctatus the was; 17.2, 19.0, 15.3, 16.7, and 16.6 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest total length was

162

22.9 cm amongst the samples collected from Chashma and Trimu Barrage and minimum total length with 11.4 cm caught from Trimu Barrage.

The of variance showed that there was non-significant difference (p > 0.05) in total length among the sites.

Table 3 ANOVA for the Total Length of C. punctatus Head Length SOV Df SS MS F-Value P-Value In case of average head length of the Sites 4 72.613 18.153 2.529 0.054 Channa punctatus the values remained Error 45 322.963 7.177 as; 3.3, 3.7, 3.0, 3.4 and 3.2 cm, respectively, from the Baloki Barrage, Total 49 395.576 Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum head length fish was 4.4 cm amongst the samples collected from Chashma Barrage and Trimu Barrage and minimum head length individuals was with 2.2 cm, caught from Trimu Barrage.

The analysis of variance showed that there was non-significant difference (p > 0.05) in head length among the sites.

Table 4 ANOVA for the Head Length of C. punctatus

SOV Df SS MS F-Value P-Value

Sites 4 2.613 0.653 2.508 0.055

Error 45 11.723 0.261

Total 49 14.336

163

Stoutness

Amongst the morphometric parameters average stoutness of the Channa punctatus remained as; 9.6, 10.6, 8.5, 9.9 and 9.3 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum stoutness was 12.7 cm amongst the samples collected from Chashma Barrage and Trimu Barrage and with minimum stoutness individuals was 6.4 cm caught from Trimu Barrage.

The data obtained for the stoutness of the Channa punctatus that there was significant difference (p ≤ 0.05) among the sites.

Table 5 ANOVA for the Stoutness of C. punctatus

SOV Df SS MS F-Value P-Value

Sites 4 22.609 5.652 2.578 0.050*

Error 45 98.676 2.193

Total 49 121.285

Dorsal Fin Length

164

Amongst the morphometric parameters average length of the dorsal fin of the Channa punctatus remained as; 1.2, 1.3, 1.1, 1.3 and 1.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the dorsal fin was 1.6 cm amongst the samples collected from Chashma Barrage and Trimu Barrage and with minimum length of the dorsal fin was 0.8 cm caught from Trimu Barrage.

The analysis of variance showed that there was non-significant difference (p > 0.05) among the sites.

Table 6 ANOVA for the Dorsal Fin Length of C. punctatus

SOV Df SS MS F-Value P-Value

Sites 4 0.305 0.76 2.096 0.097

Error 45 1.636 0.36

Total 49 1.941

Caudal Fin Length

Amongst the morphometric parameters average length of the caudal fin of the Channa punctatus remained as; 1.9, 2.1, 1.7, 2.0 and 2.0 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the caudal fin was 2.6 cm amongst the samples collected from Trimu Barrage and with minimum length of the caudal fin individuals was 1.3 cm caught from Trimu Barrage.

The analysis of variance showed that there was significant difference (p < 0.05) among the sites.

Table 7 ANOVA for the Caudal Fin Length of C. punctatus

165

SOV Df SS MS F-Value P-Value

Sites 4 .940 .235 2.64452 0.046*

Error 45 4.000 .089

Total 49 49.19

Average Length of Paired Pectoral Fins

Amongst the morphometric parameters average length of the paired pectoral fin of the Channa punctatus remained as; 1.9, 2.1, 1.7, 2.0 and 2.0 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum and minimum length of the paired pectoral fin was 2.6 cm and 1.3 cm, respectively, amongst the samples collected from Trimu Barrage.

The analysis of variance showed that there was significant difference (p < 0.05) among the sites.

Table 8 ANOVA for the Average Length of Paired Pectoral Fins of C. punctatus

SOV Df SS MS F-Value P-Value

Sites 4 .940 .235 2.644 0.046*

Error 45 4.000 .089

Total 49 49.19

B. Morphometric Parameters of and Channa marulius

166

Body Weight

The average body weight of the fish C. marulius remained as; 2026.2, 2180.3, 1852.2, 1875.6 and 1881.7 g, respectively, collected from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest weight fish was 3000 g amongst samples collected from Chashma Barrage and minimum weight was from Baloki Barrage with 1000 g.

The data for the body weight of C. marulius showed that there was non-significant difference (p > 0.05) in total length among the sites.

Table 9 ANOVA for the Wet Body Weight of Channa marulius

SOV Df SS MS F-Value P-Value

Sites 4 777384 194346 0.51 0.731

Error 45 17280820 384018

Total 49 18058204

Total Length

In case of average total length of the C. marulius the values remained as; 48.4, 52.1, 44.9, 44.2 and 44.8 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest total length fish was 71.6 cm amongst the samples collected from Chashma Barrage and minimum total length fish was with 23.9 cm caught from Baloki Barrage.

The analysis of variance showed that there was non-significant difference (p > 0.05) among the sites

167

Table 10 ANOVA for the Total Length of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 443 111 0.51 0.731

Error 45 9851 219

Total 49 10294

Head Length In case of average head length of the C. marulius the values remained as; 11.3, 12.2, 10.5, 10.4 and 10.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum head length fish was with 16.8 cm amongst the samples collected from Chashma Barrages and minimum head length was 5.6 cm caught from Baloki Barrage.

The analysis of variance showed that there was non-significant difference (p > 0.05) among the sites.

Table 11 ANOVA for the Head Length of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 24.3 6.1 0.51 0.731

Error 45 539.6 12.0

Total 49 563.9

168

Stoutness

Amongst the morphometric parameters average stoutness of the C. marulius remained as; 22.1, 23.8, 20.6, 20.2 and 20.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum stoutness fish was with 32.8 cm amongst the samples collected from Chashma Barrage and with minimum stoutness was 10.9 cm caught from Baloki Barrage.

The analysis of variance showed that there was non-significant difference (p > 0.05) among the sites

Table 12 ANOVA for the Stoutness of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 92.7 23.2 0.51 0.731

Error 45 2061.4 45.8

Total 49 2154.2

Dorsal Fin Length

Amongst the morphometric parameters average length of the dorsal fin of the C. marulius remained as; 3.1, 3.3, 2.9, 2.8 and 2.9 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the dorsal fin holding individual fish was with 4.6 cm amongst the samples collected from Chashma Barrages and with minimum length of the dorsal fin individual was 1.5 cm caught from Baloki Barrage.

The data for the length of the dorsal fin of the C. marulius showed that there was non-significant difference (p > 0.05) among the sites.

169

Table 13 ANOVA for the Dorsal Fin Length of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 1.806 0.451 0.51 0.823

Error 45 40.136 0.892

Total 49 41.942

Caudal Fin Length

Amongst the morphometric parameters average length of the caudal fin of the C. marulius remained as; 8.2, 8.9, 7.6, 7.5 and 7.6 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the caudal fin fish was with 12.2 cm amongst the samples collected from Chashma Barrages and with minimum length of the caudal fin individual was 4.1 cm caught from Baloki Barrage.

The data for the length of the caudal fin of the C. marulius showed that there was non-significant difference (p > 0.05) among the sites.

Table 14 ANOVA for the Caudal Fin Length of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 12.84 3.21 0.51 0.731

Error 45 285.41 6.34

Total 49 298.25

170

Anal Fin Length

Amongst the morphometric parameters average length of the anal fin of the C. marulius remained as; 3.1, 3.3, 2.9, 2.8 and 2.9 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the anal fin fish was with 4.6 cm amongst the samples collected from Chashma Barrages and with minimum length of the anal fin individual was 1.5 cm caught from Baloki Barrage.

The data for the length of the anal fin of the C. marulius showed that there was non-significant difference (p > 0.05) among the sites.

Table 15 ANOVA for the Anal Fin Length of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 1.806 0.451 0.51 0.731

Error 45 40.136 0.892

Total 49 41.942

Average Length of Paired Pectoral Fins

Amongst the morphometric parameters average length of paired pectoral fins of the C. marulius remained as; 7.2, 7.8, 6.7, 6.6 and 6.7 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pectoral fins individual fish was with 10.7 cm amongst the samples collected from Chashma Barrages and with minimum average length of paired pectoral fins individual was 3.6 cm each caught from Baloki Barrage.

171

The data for the average length of paired pectoral fins of the C. marulius showed that there was non-significant difference (p > 0.05) among different experimental sites.

Table 16 ANOVA for the Average Length of Paired Pectoral Fins of C. marulius

SOV Df SS MS F-Value P-Value

Sites 4 9.83 2.46 0.51 0.731

Error 45 218.52 4.86

Total 49 228.35

C. Morphometric Parameters of Rita rita

Body Weight

The average body weight of the R. rita remained as; 179.3, 149.2, 152.6, 156.1 and 162.7 g, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest weight was 233 g amongst samples collected from Baloki Barrage and minimum weight individual was from Chashma Barrage with 102 g.

The data analyzed for the body weight of R. rita showed that there was non-significant difference (p> 0.05) among the sites.

172

Table 17 ANOVA for the Wet Body Weight of Rita rita

SOV Df SS MS F-Value P-Value

Sites 4 5664 1416 1.61 0.188

Error 45 39553 879

Total 49 45217

Fork Length

In case of average fork length of the R. rita the values remained as, 25.8, 21.5, 22.0, 22.5 and 23.4 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest fork length holding was 33.6 cm amongst the samples collected from Baloki Barrage and minimum total fork was with 14.7 cm caught from Chashma Barrage.

The data analyzed for the fork length showed that there was non-significant difference (p > 0.05) among the sites.

Table 18 ANOVA for the Fork Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 117.5 29.4 1.61 0.188

Error 45 820.4 18.2

Total 49 937.9

173

Total Length

In case of average total length of the R. rita the values remained as; 30.3, 25.2, 25.8, 26.4 and 27.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest total length of fish was 39.4 cm amongst the samples collected from Baloki Barrage and minimum total length individual was with 17.3 cm caught from Chashma Barrage.

The data for the total length showed that there was non-significant difference (p > 0.05) among the sites.

Table 19 ANOVA for the Total Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 162.1 40.5 1.61 0.188

Error 45 1131.9 25.2

Total 49 1293.9

Head Length

In case of average head length of the R. rita the values remained as; 4.6, 3.8, 3.9, 4.0 and 4.1 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum head length was with 5.9 cm amongst the samples collected from Baloki Barrage and minimum head length was 2.6 cm caught from Chashma Barrage.

The data obtained head length showed that there was non-significant difference (p > 0.05) among the sites.

174

Table 20 ANOVA for the Head Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 3.654 0.914 1.61 0.188

Error 45 25.518 0.567

Total 49 29.172

Stoutness

Amongst the morphometric parameters average stoutness of the R. rita remained as; 30.4, 25.3, 25.9, 26.4 and 27.6 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum stoutness was 39.5 cm amongst the samples collected from Baloki Barrage and with minimum stoutness was 17.3 cm caught from Chashma Barrage.

The data for the stoutness of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 21 ANOVA for the Stoutness of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 162.6 40.6 1.61 0.188

Error 45 1135.3 25.2

Total 49 1297.8

175

Dorsal Fin Length

Amongst the morphometric parameters average length of the dorsal fin of the R. rita remained as; 6.8, 6.7, 5.8, 5.9 and 6.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the dorsal fin was 8.9 cm amongst the samples collected from Baloki Barrage and with minimum length of the dorsal fin individual was 3.9 cm caught from Chashma Barrage.

The data for the length of the dorsal fin of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 22 ANOVA for the Dorsal Fin Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 8.22 2.06 1.61 0.188

Error 45 57.42 1.28

Total 49 65.64

Caudal Fin Length

Amongst the morphometric parameters average length of the caudal fin of the R. rita remained as; 6.1, 5.0, 5.2, 5.3 and 5.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the caudal fin was 7.9 cm amongst the samples collected from Baloki Barrage and with minimum length of the caudal fin individual was 3.4 cm caught from Chashma Barrage.

176

The data for the length of the caudal fin of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 23 ANOVA for the Caudal Fin Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 6.46 1.62 1.61 0.188

Error 45 45.14 1.00

Total 49 51.60

Anal Fin Length

Amongst the morphometric parameters average length of the anal fin of the R. rita remained as; 3.1, 2.5, 2.6, 2.7 and 2.8 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the anal fin was 4.0 cm amongst the samples collected from Baloki Barrage and with minimum length of the anal fin was 1.7 cm caught from Chashma Barrage.

The data for the length of the anal fin of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 24 ANOVA for the Anal Fin Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 1.640 0.410 1.61 0.188

177

Error 45 11.455 0.255

Total 49 13.095

Adipose Fin Length

Amongst the morphometric parameters average length of the adipose fin of the R. rita remained as; 1.2, 1.0, 1.0, 1.0 and 1.1 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the adipose fin holding individual fish was with 1.5 cm amongst the samples collected from Baloki Barrage and with minimum length of the adipose fin was 0.7 cm, one each, caught from Chashma and Trimu Barrages.

The data for the length of the adipose fin of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 25 ANOVA for the Adipose Fin Length of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 0.2470 0.0618 1.61 0.188

Error 45 1.7250 0.0383

Total 49 1.9720

178

Average Length of Paired Pectoral Fins

Amongst the morphometric parameters average length of paired pectoral fins of the R. rita remained as; 6.1, 5.0, 5.2, 5.3 and 5.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pectoral fins holding fish was with 7.9 cm amongst the samples collected from Baloki Barrage and with minimum average length of paired pectoral fins was 3.4 cm each caught from Chashma and Trimu Barrages.

The data for the average length of paired pectoral fins of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 26 ANOVA for the Average Length of Paired Pectoral Fins of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 6.46 1.62 1.61 0.188

Error 45 45.14 1.00

Total 49 51.60

Average Length of Paired Pelvic Fins

Amongst the morphometric parameters average length of paired pelvic fins of the R. rita remained as; 2.4, 2.0, 2.1, 2.1 and 2.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pelvic fins was 3.2 cm amongst the samples collected from Baloki Barrage and with minimum average length of paired pelvic fins was 1.4 cm caught from Chashma Barrage.

179

The data for the average length of paired pelvic fins of the R. rita showed that there was non-significant difference (p > 0.05) among the sites.

Table 27 ANOVA for the Average Length of Paired Pelvic Fins of R. rita

SOV Df SS MS F-Value P-Value

Sites 4 1.038 0.260 1.61 0.188

Error 45 7.249 0.161

Total 49 8.287

D. Morphometric Parameters of Sperata seenghala

Body Weight

The average body weight of the S. seenghala remained as; 1522.5 1942.5, 1558.9, 2293.5 and 2089.0 g, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest weight having individual fish was 3000 g amongst samples collected from Trimu Barrage and minimum weight was from Qadirabad and Baloki Barrages with 1200 g.

180

The data for the body weight of S. seenghala showed that there was highly significant difference (p < 0.01) among the sites.

Table 28 ANOVA for the Wet Body Weight of Sperata seenghala

SOV Df SS MS F-Value P-Value

Sites 4 4494728 1123682 8.27 0.000**

Error 45 6117306 135940

Total 49 10612034

Fork Length

In case of average fork length of the S. seenghala the values remained as; 30.2, 38.5, 30.9, 45.4 and 41.4 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest fork length was 59.4 cm amongst the samples collected from Trimu Barrage and minimum total fork individuals were with 23.8 cm caught from Qadirabad and Baloki Barrages.

The data for the fork length showed that there was highly significant difference (p < 0.05) among the sites.

Table 29 Analysis of variance on the Fork Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 1764.2 441.1 8.27 0.000**

181

Error 45 2401.1 53.4

Total 49 4165.4

Total Length

In case of average total length of the S. seenghala the values remained as; 37.5, 47.9, 38.4, 56.5 and 51.5 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest total length having individual fish was 73.9 cm amongst the samples collected from Trimu Barrage and minimum total length individuals were with 29.6 cm caught from Qadirabad and Baloki Barrages.

The data for the total length showed highly significant difference (p < 0.01) among the sites.

Table 30 ANOVA for the Total Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 2728.4 682.1 8.27 0.000**

Error 45 3713.4 82.5

Total 49 6441.8

Head Length

In case of average head length of the S. seenghala the values remained as; 8.5, 10.9, 8.7, 12.8 and 11.7 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The

182

maximum head length was 16.8 cm amongst the samples collected from Trimu Barrage and minimum head length individuals were with 7 cm caught from Qadirabad and Baloki Barrages.

The data head length showed highly significant difference (p < 0.01) among the sites.

Table 31 ANOVA for the Head Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 140.35 35.09 8.27 0.000**

Error 45 191.02 4.24

Total 49 331.37

Stoutness

Amongst the morphometric parameters average stoutness of the S. seenghala remained as; 15.9, 20.2, 16.2, 23.9 and 21.8 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum stoutness was 31.2 cm amongst the samples collected from Trimu Barrage and with minimum stoutness was 12.5 cm caught from Qadirabad and Baloki Barrages.

The data for the stoutness of the S. seenghala showed highly significant difference (p < 0.01) among the sites.

Table 32 ANOVA for the Stoutness of S. seenghala

SOV Df SS MS F-Value P-Value

183

Sites 4 487.5 121.9 8.27 0.000**

Error 45 663.4 14.7

Total 49 1150.9

Dorsal Fin Length

Amongst the morphometric parameters average length of the dorsal fin of the S. seenghala remained as; 5.1, 6.4, 5.2, 7.6 and 6.9 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the dorsal fin was 10.0 cm amongst the samples collected from Trimu Barrage and with minimum length of the dorsal fin was 4.0 cm one each caught from Qadirabad and Baloki Barrages Barrage.

The data for the length of the dorsal fin of the S. seenghala highly significant difference (p < 0.01) among the sites.

Table 33 ANOVA for the Dorsal Fin Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 49.46 12.37 8.27 0.000**

Error 45 67.32 1.50

Total 49 116.78

184

Caudal Fin Length

Amongst the morphometric parameters average length of the caudal fin of the S. seenghala remained as; 7.3, 9.4, 7.5, 11.1 and 10.1 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the caudal fin was 14.5 cm amongst the samples collected from Trimu Barrage and with minimum length of the caudal fin was 5.8 cm caught from Qadirabad and Baloki Barrages.

The data for the length of the caudal fin of the S. seenghala showed highly significant difference (p < 0.01) among the sites.

Table 34 ANOVA for the Caudal Fin Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 104.68 26.17 8.27 0.000**

Error 45 142.47 3.17

Total 49 247.16

185

Anal Fin Length

Amongst the morphometric parameters average length of the anal fin of the S. seenghala remained as; 3.1, 3.9, 3.2, 4.7 and 4.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the anal fin holding was 6.1 cm amongst the samples collected from Trimu Barrage and with minimum length of the anal fin was 2.4 cm caught from Qadirabad, Baloki and Trimu Barrages.

The data for the length of the anal fin of the S. seenghala showed highly significant difference (p < 0.01) among the sites.

Table 35 ANOVA for the Anal Fin Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 18.559 4.640 8.27 0.000**

Error 45 25.259 0.561

Total 49 43.817

Adipose Fin Length

Amongst the morphometric parameters average length of the adipose fin of the S. seenghala remained as 1.9, 2.5, 2.0, 2.9 and 2.7 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the adipose fin was 3.8 cm amongst the samples collected from Trimu Barrage and with minimum length of the adipose fin was 1.5 cm one each, caught from Qadirabad and Baloki Barrages.

The data for the length of the adipose fin of the S. seenghala showed highly significant difference (p < 0.01) among the sites. 186

Table 36 ANOVA for the Adipose Fin Length of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 7.250 1.812 8.27 0.000***

Error 45 9.867 0.219

Total 49 17.116

Average Length of Paired Pectoral Fins

Amongst the morphometric parameters average length of paired pectoral fins of the S. seenghala remained as; 3.5, 4.4, 3.6, 5.2 and 4.8 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pectoral fins was 6.9 cm amongst the samples collected from Trimu Barrage and with minimum average length of paired pectoral fins was 2.7 cm one each, caught from Qadirabad and Baloki Barrages.

The data for the average length of paired pectoral fins of the S. seenghala showed highly significant difference (p < 0.01) among the sites.

Table 37 ANOVA for the Av. Length of Paired Pectoral Fins of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 23.489 5.872 8.27 0.000**

Error 45 31.968 0.710

187

Total 49 55.456

Average Length of Paired Pelvic Fins

Amongst the morphometric parameters average length of paired pelvic fins of the S. seenghala remained as; 3.1, 3.2, 3.2, 4.7 and 4.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pelvic fins was 6.9 cm amongst the samples collected from Trimu Barrage and with minimum average length of paired pelvic fins was 2.4 cm, one each, caught from Baloki, Chashma and Qadirabad Barrages.

The data for the average length of paired pelvic fins of the S. seenghala showed highly significant difference (p < 0.01) among the sites.

Table 38 ANOVA for the Av. Length of Paired Pelvic Fins of S. seenghala

SOV Df SS MS F-Value P-Value

Sites 4 21.480 5.370 7.64 0.000**

Error 45 24.730 0.550

Total 49 46.210

E. Morphometric Parameters of Wallago attu

188

Body Weight

The average body weight of the collected fish W. attu remained as; 1995.2, 2025.3, 1924.6, 2348.3 and 1717.5 g, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest weight was 2980 g amongst samples collected from Trimu Barrage and minimum weight individual was from Taunsa Barrage with 1155 g.

The data for the body weight of W. attu showed that there was non-significant difference (p > 0.05) among the sites.

Table 39 ANOVA for the Wet Body Weight of Wallago attu

SOV Df SS MS F-Value P-Value

Sites 4 2074437 518609 1.71 0.164

Error 45 13625643 302792

Total 49 15700079

Fork Length

In case of average fork length of the W. attu the values remained as; 46.1, 46.8, 44.5, 54.3 and 39.7 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest fork length having individual fish was 68.9 cm amongst the samples collected from Trimu Barrage and minimum total fork was with 26.7 cm caught from Chashma Barrage.

189

The data obtained for the fork length was subjected to Minitab computer software and the analysis of variance showed that there was non-significant difference (p > 0.05) among the sites.

Table 40 ANOVA for the Fork Length of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 1108 277 1.71 0.164

Error 45 7280 162

Total 49 8388

Total Length

In case of average total length of the W. attu the values remained as; 50.2, 50.9, 48.4, 59.1 and 43.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The highest total length was 74.9 cm amongst the samples collected from Trimu Barrage and minimum total length was with 29.0 cm caught from Chashma Barrage.

The data for the total length showed that there was non-significant difference (p > 0.05) among the sites.

190

Table 41 ANOVA for the Total Length of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 1312 328 1.71 0.164

Error 45 8616 191

Total 49 9927

Head Length

In case of average head length of the W. attu the values remained as; 11.1, 11.3, 10.8, 13.1 and 9.6 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum head length was 16.7 cm amongst the samples collected from Trimu Barrage and minimum head length was 6.5 cm caught from Taunsa Barrage.

The data head length showed that there was non-significant difference (p > 0.05) among the sites.

Table 42 ANOVA for the Head Length of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 64.78 16.19 1.71 0.164

Error 45 425.47 9.45

Total 49 490.25

191

Stoutness

Amongst the morphometric parameters average stoutness of the W. attu remained as; 18.2, 18.5, 17.6, 21.5 and 15.7 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum stoutness was in two fish samples with 27.2 cm amongst the samples collected from Trimu Barrage and with minimum stoutness was 10.6 cm caught from Chashma Barrage.

The data for the stoutness of the W. attu showed that there was non-significant difference (p > 0.05) among the sites.

Table 43 ANOVA for the Stoutness of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 173.4 43.4 1.71 0.164

Error 45 1139.3 25.3

Total 49 1312.7

Dorsal Fin Length

Amongst the morphometric parameters average length of the dorsal fin of the W. attu remained as; 6.1, 6.2, 5.9, 7.2 and 5.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum length of the dorsal fin was in two fish samples with 9.1 cm amongst the samples collected from Trimu Barrage and with minimum length of the dorsal fin was 3.5 cm caught from Taunsa Barrage.

The data for the length of the dorsal fin of the W. attu showed that there was non-significant difference (p > 0.05) among the sites.

192

Table 44 ANOVA for the Dorsal Fin Length of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 19.27 4.82 1.71 0.164

Error 45 126.59 2.81

Total 49 145.86

Caudal Fin Length

Amongst the morphometric parameters average length of the caudal fin of the W. attu remained as; 4.6, 4.6, 4.4, 5.4 and 3.9 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pelvic fins was in two fish samples with 6.8 cm amongst the samples collected from Trimu Barrage and with minimum average length of paired pelvic fins individual was 2.6 cm Taunsa Barrage.

The data for the length of the caudal fin of the W. attu showed that there was non-significant difference (p > 0.05) among the sites.

Table 45 ANOVA for the Caudal Fin Length of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 10.84 2.71 1.71 0.164

Error 45 71.20 1.58

193

Total 49 82.05

Average Length of Paired Pectoral Fins

Amongst the morphometric parameters average length of paired pectoral fins of the W. attu remained as; 6.1, 6.2, 5.9, 7.2 and 5.2 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pelvic fins was in two fish samples with 9.1 cm amongst the samples collected from Trimu Barrage and with minimum average length of paired pelvic fins was 3.5 cm Taunsa Barrage.

The data for the average length of paired pectoral fins of the W. attu showed that there was non-significant difference (p > 0.05) among the sites.

Table 46 ANOVA for the Average Length of Paired Pectoral Fins of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 19.27 4.82 1.71 0.164

Error 45 126.59 2.81

Total 49 145.86

Average Length of Paired Pelvic Fins

Amongst the morphometric parameters average length of paired pelvic fins of the W. attu remained as; 3.0, 3.1, 2.9, 3.6 and 2.6 cm, respectively, from the Baloki Barrage, Chashma Barrage, Qadirabad Barrage, Trimu Barrage and Taunsa Barrage. The maximum average length of paired pelvic fins holding fish samples was with 4.5 cm amongst the samples

194

collected from Trimu Barrage and one from Chashma barrage and with minimum average length of paired pelvic fins was 1.8 cm, one each, caught from Chashma and Taunsa Barrages.

The data for the average length of paired pelvic fins of the W. attu showed that there was non-significant difference (p > 0.05) among the sites

Table 47 ANOVA for the Average Length of Paired Pelvic Fins of W. attu

SOV Df SS MS F-Value P-Value

Sites 4 4.818 1.205 1.71 0.164

Error 45 31.647 07.3

Total 49 36.465

4.2 Principal Component Analysis for Morphometric Parameters of Experimental Species

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among samples of all fish species selected collected from different geographical locations along with differentiation between and within the groups.

A Principle Component Analysis (PCA) of Channa punctatus

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among the Channa punctatus collected samples from different geographical locations along with differentiation between and within the groups

195

The summary statistics, correlation matrix (Pearson (n)) of PCA and Bartlett’s sphericity test of PCA are shown in Table 48, Table 49 and Table 50, respectively. The results obtained from the PCA indicated clearly that the augment in the number of factors or components was correlated with the decrease in Eigen values (Figure 7). The values showed that its trend reached its maximum at level of first factor (Table 51). In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first two main factors accounted for 98.706% of cumulative variability. Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 98.71% of the cumulative variation among the morphometric parameters of study (Figure 8). The first group amongst the major two groups accounted for 84.415% of the variability while the second accounted for 14.291% of the cumulative variability. The variability percentage for factor F3 and F4, the values were 0.825% and 0.469%, respectively (Table 51). The correlation of F1 and F2 for all the parameters was positive except for dorsal fin for F2. F3 and F4 were positive in correlation for wet body weight, total length, dorsal fin length and for wet body weight, total length, dorsal fin length average length of paired pectoral fin, respectively (Table 54).

The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups (Figures 9 and 10). On the observation of the these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on observation plot analysis (Figure 9) and the F1 and F2 factors bi-plot (Figure 10). The Eigen vectors, factor loading values and the percentage contribution of the variables are presented (Tables 52, 53 and 55). The bold values in the squared cosines of the variables showed, which are the major two factors on the basis of which the parameters were studied (Table 56).

196

Table 48 Summary statistics of PCA for C. punctatus

Variable Obser. Obs. with Obs. without Min. Max. Mean Std. deviation missing data missing data Wt (g) 50 0 50 90.000 180.000 135.240 22.334

TL (cm) 50 0 50 11.430 22.860 17.175 2.836

HL (cm) 50 0 50 2.217 4.445 3.332 0.551

St (cm) 50 0 50 6.355 12.710 9.549 1.577

DF (cm) 50 0 50 0.823 1.646 1.229 0.196

CF (cm) 50 0 50 1.280 2.560 1.923 0.317

PF (cm) 50 0 50 1.300 2.540 1.908 0.330

Table 49 Correlation matrix (Pearson (n)) of PCA for C. punctatus

Variables Wt TL HL St DF CF PF

197

Wt 1 0.018 0.018 0.018 -0.007 0.017 0.023

TL 0.018 1 1.000 1.000 0.966 1.000 0.979

HL 0.018 1.000 1 1.000 0.965 1.000 0.979

St 0.018 1.000 1.000 1 0.966 1.000 0.979

DF -0.007 0.966 0.965 0.966 1 0.968 0.947

CF 0.017 1.000 1.000 1.000 0.968 1 0.979

PF 0.023 0.979 0.979 0.979 0.947 0.979 1

Values in bold are different from 0 with a significance level alpha=0.05

Table 50 Bartlett's sphericity test of PCA for C. punctatus

Chi-square (Observed value) -Inf

Chi-square (Critical value) 32.671

DF 21

p-value -

Alpha 0.05

Table 51 Eigen values of PCA for C. punctatus

198

F1 F2 F3 F4

Eigenvalue 5.909 1.000 0.058 0.033

Variability (%) 84.415 14.291 0.825 0.469

Cumulative % 84.415 98.706 99.531 100.000

Figure 7 Scree plot between Eigen values, factors and cumulative variability for C. punctatus

199

Table 52 Eigen vectors of PCA for C. punctatus

F1 F2 F3 F4

Wt 0.007 1.000 0.025 0.001

TL 0.411 0.001 -0.133 -0.259

HL 0.411 0.001 -0.143 -0.263

St 0.411 0.001 -0.130 -0.257

DF 0.401 -0.026 0.892 0.203

CF 0.411 0.000 -0.099 -0.243

PF 0.405 0.006 -0.372 0.835

Table 53 Factor loadings of PCA for C. punctatus

F1 F2 F3 F4

Wt 0.017 1.000 0.006 0.000

TL 0.998 0.001 -0.032 -0.047

200

HL 0.998 0.001 -0.034 -0.048

St 0.998 0.001 -0.031 -0.047

DF 0.976 -0.026 0.214 0.037

CF 0.999 0.000 -0.024 -0.044

PF 0.984 0.006 -0.089 0.151

Table 54 Correlations between variables and factors of PCA for C. punctatus

F1 F2 F3 F4

Wt 0.017 1.000 0.006 0.000

TL 0.998 0.001 -0.032 -0.047

201

HL 0.998 0.001 -0.034 -0.048

St 0.998 0.001 -0.031 -0.047

DF 0.976 -0.026 0.214 0.037

CF 0.999 0.000 -0.024 -0.044

PF 0.984 0.006 -0.089 0.151

Figure 8 Graph between variables for C. punctatus

202

Table 55 Contribution of the variables (%) of PCA for C. punctatus

F1 F2 F3 F4

Wt 0.005 99.930 0.065 0.000

TL 16.869 0.000 1.774 6.686

HL 16.865 0.000 2.045 6.913

St 16.870 0.000 1.684 6.608

DF 16.112 0.066 79.584 4.134

CF 16.881 0.000 0.983 5.927

PF 16.399 0.004 13.865 69.732

Table 56 Squared cosines of the variables of PCA for C. punctatus

F1 F2 F3 F4

203

Wt 0.000 1.000 0.000 0.000

TL 0.997 0.000 0.001 0.002

HL 0.997 0.000 0.001 0.002

St 0.997 0.000 0.001 0.002

DF 0.952 0.001 0.046 0.001

CF 0.997 0.000 0.001 0.002

PF 0.969 0.000 0.008 0.023

Values in bold correspond for each variable to the factor for which the squared cosine is the largest

Figure 9 Graph between Observations for C. punctatus

204

Figure 10 Biplot graph between factors for C. punctatus

205

B Principle Component Analysis (PCA) of Channa marulius

206

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among the C. marulius collected samples from different geographical locations along with differentiation between and within the groups.

The summary statistics, correlation matrix (Pearson (n)) of PCA and Bartlett’s sphericity test of PCA are shown in Table 57, Table 58 and Table 59, respectively. The results obtained from the PCA indicated clearly that the augment in the number of factors or components was correlated with the decrease in Eigen values (Figure 11). The values showed that its trend reached its maximum at level of second factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first two main factors accounted for 99.996% of cumulative variability. Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 100% of the cumulative variation among the morphometric parameters of study (Figure 12). The first group amongst the major two groups accounted for 99.849% of the cumulative variability while the second accounted for 0.148% of the cumulative variability (Table 60). Factor three (F3) has a value of 0.003% only. The factors one (F1) had a positive correlation for all the factors whereas the factor two (F2) were negatively correlated with all the parameters except dorsal fin length. The other factors F4, F5 and F6 had a zero or negative value except for stoutness and Head length for F4 and F5 respectively, where the correlation is positive (Table 63).

The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups (Figures13 and 14). On the observation of these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 factors bi-plot (Figure 14) and observation plot analysis (Figure 13). The Eigen vectors, factor loading values and the percentage contribution of the variables are presented (Tables 61, 62, and 64). The bold values in the

207

squared cosines of the variables showed, which are the major two factors on the basis of which the parameters were studied (Table 65).

Table 57 Summary statistics for C. marulius

Variable Obser Obs. with missing Obs. without Min Max Mean Std. deviation data missing data

Wt (g) 50 0 50 1000.00 3000.00 1963.20 607.07

TL (cm) 50 0 50 23.876 71.628 46.876 14.497

HL (cm) 50 0 50 5.588 16.764 10.972 3.391

St (cm) 50 0 50 10.922 32.766 21.450 6.638

DF (cm) 50 0 50 1.524 4.572 3.008 0.914

CF (cm) 50 0 50 4.364 12.192 7.984 2.458

AF (cm) 50 0 50 1.524 4.572 2.992 0.925

PF (cm) 50 0 50 3.556 10.668 6.981 2.159

Table 58 Correlation matrix (Pearson (n)) for C. marulius

Variables Wt TL HL St DF CF AF PF

Wt 1 1.000 1.000 1.000 0.993 1.000 1.000 1.000

208

TL 1.000 1 1.000 1.000 0.993 1.000 1.000 1.000

HL 1.000 1.000 1 1.000 0.993 1.000 1.000 1.000

St 1.000 1.000 1.000 1 0.993 1.000 1.000 1.000

DF 0.993 0.993 0.993 0.993 1 0.993 0.993 0.993

CF 1.000 1.000 1.000 1.000 0.993 1 1.000 1.000

AF 1.000 1.000 1.000 1.000 0.993 1.000 1 1.000

PF 1.000 1.000 1.000 1.000 0.993 1.000 1.000 1

Values in bold are different from 0 with a significance level alpha=0.05

Table 59 Bartlett’s sphericity test for C. marulius

Chi-square (Observed -Inf value) Chi-square (Critical 41.337 value) DF 28

p-value

Alpha 0.05

Table 60 Eigen values of PCA for C. marulius

209

F1 F2 F3 F4 F5 F6

Eigenvalue 7.988 0.012 0.000 0.000 0.000 0.000

Variability (%) 99.849 0.148 0.003 0.000 0.000 0.000

Cumulative % 99.849 99.996 99.999 100.000 100.000 100.000

Figure 11 Scree plot between Eigen values, factors and cumulative variability for C. marulius

Table 61 Eigen vectors of PCA for C. marulius 210

F1 F2 F3 F4 F5 F6

Wt 0.354 -0.131 -0.149 -0.167 -0.227 -0.299

TL 0.354 -0.131 -0.148 -0.182 -0.267 0.855

HL 0.354 -0.130 -0.166 -0.243 0.878 0.027

St 0.354 -0.132 -0.174 0.908 0.056 0.014

DF 0.352 0.936 0.011 0.001 -0.001 0.000

CF 0.354 -0.144 0.924 0.018 0.016 0.000

AF 0.354 -0.131 -0.149 -0.167 -0.227 -0.299

PF 0.354 -0.131 -0.149 -0.167 -0.227 -0.299

Table 62 Factor loadings of PCA for C. marulius

F1 F2 F3 F4 F5 F6

Wt 1.000 -0.014 -0.002 -0.001 -0.001 0.000

TL 1.000 -0.014 -0.002 -0.001 -0.001 0.001

211

HL 1.000 -0.014 -0.003 -0.001 0.003 0.000

St 1.000 -0.014 -0.003 0.006 0.000 0.000

DF 0.995 0.102 0.000 0.000 0.000 0.000

CF 1.000 -0.016 0.014 0.000 0.000 0.000

AF 1.000 -0.014 -0.002 -0.001 -0.001 0.000

PF 1.000 -0.014 -0.002 -0.001 -0.001 0.000

Table 63 Correlations between variables and factors of PCA for C. marulius

F1 F2 F3 F4 F5 F6

Wt 1.000 -0.014 -0.002 -0.001 -0.001 0.000

TL 1.000 -0.014 -0.002 -0.001 -0.001 0.001

HL 1.000 -0.014 -0.003 -0.001 0.003 0.000

St 1.000 -0.014 -0.003 0.006 0.000 0.000

DF 0.995 0.102 0.000 0.000 0.000 0.000

CF 1.000 -0.016 0.014 0.000 0.000 0.000

212

AF 1.000 -0.014 -0.002 -0.001 -0.001 0.000

PF 1.000 -0.014 -0.002 -0.001 -0.001 0.000

Figure 12 Graph between variables for C. marulius

213

Table 64 Contribution of the variables (%) of PCA for C. marulius

F1 F2 F3 F4 F5 F6

Wt 12.516 1.723 2.215 2.786 5.171 8.923

TL 12.516 1.722 2.191 3.327 7.103 73.141

HL 12.516 1.696 2.763 5.904 77.050 0.071

St 12.516 1.737 3.035 82.381 0.310 0.020

DF 12.389 87.598 0.013 0.000 0.000 0.000

CF 12.513 2.078 85.353 0.031 0.025 0.000

AF 12.516 1.723 2.215 2.786 5.171 8.923

PF 12.516 1.723 2.215 2.786 5.171 8.923

Table 65 Squared cosines of the variables of PCA for C. marulius

F1 F2 F3 F4 F5 F6

Wt 1.000 0.000 0.000 0.000 0.000 0.000

TL 1.000 0.000 0.000 0.000 0.000 0.000

HL 1.000 0.000 0.000 0.000 0.000 0.000

214

St 1.000 0.000 0.000 0.000 0.000 0.000

DF 0.990 0.010 0.000 0.000 0.000 0.000

CF 1.000 0.000 0.000 0.000 0.000 0.000

AF 1.000 0.000 0.000 0.000 0.000 0.000

PF 1.000 0.000 0.000 0.000 0.000 0.000 Values in bold correspond for each variable to the factor for which the squared cosine is the largest

Figure 13 Graph between Observations for C. marulius

215

Figure 14 Biplot graph between factors for C. marulius

216

C Principle Component Analysis (PCA) of Rita rita

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among the Rita rita collected samples from different geographical locations along with differentiation between and within the groups.

217

The summary statistics, correlation matrix (Pearson (n)) of PCA and Bartlett’s sphericity test of PCA are shown in Table 66, Table 67 and Table 68, respectively. The results obtained from the PCA indicated clearly that the augment in the number of factors or components was correlated with the decrease in Eigen values (Figure 15). The values showed that its trend reached its maximum at level of first factor (Table 69). In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first two main factors accounted for 99.016% of cumulative variability. Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 99.02% of the cumulative variation among the morphometric parameters of study (Figure16). The first group amongst the major two groups accounted for 96.85% of the cumulative variability while the second accounted for 2.16% of the cumulative variability. The factor one was positively correlated for all noted body parameters and factor two (F2) were negatively correlated with all the morphometric characters measured except for the adipose fin length for the Rita rita. F3 was negatively correlated except for anal fin, F4 was positively correlated for dorsal fin length, anal fin length and adipose fin length, F5 had a positive correlation for head length dorsal fin length, adipose fin length while other parameters had a negative correlation ship, F6 had a positive correlation for head length and average length paired pectoral fin while negative for other observed parameters, F7 had a positive correlation for caudal fin and a negative for total length and average length paired pectoral fin and F8 were negatively correlated for body weight, fork length, stoutness and average length of the paired pelvic fins (Table 72).

The graphical presentation shows the correlation amongst the loadings and factors by PCA (Figure16). The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups (Figures 17 and 18). On the observation of the these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 factors bi-plot (Figure 18) and observation plot analysis (Figure 17). The Eigen vectors, factor loading values and the percentage contribution of the factors are presented (Tables 70, 71 and 73). The bold values in the squared

218

cosines of the variables showed, which are the major two factors on the basis of which the parameters were studied (Table 74).

Table 66 Summary statistics for R. rita

Variable Obs Obs. with Obs. without Min. Max. Mean Std. missing missing data deviation data Wt (g) 50 0 50 102.000 233.000 159.980 30.378

FL (cm) 50 0 50 14.690 33.556 23.040 4.375

TL (cm) 50 0 50 17.655 39.415 27.073 5.126

HL (cm) 50 0 50 2.591 5.918 4.065 0.770

St (cm) 50 0 50 17.281 39.474 27.103 5.147

DF (cm) 50 0 50 3.886 8.877 6.128 1.139

CF (cm) 50 0 50 3.446 7.871 5.406 1.026

AF (cm) 50 0 50 1.736 3.995 2.765 0.553

AdF (cm) 50 0 50 0.674 1.546 1.081 0.220

PF (cm) 50 0 50 3.446 7.871 5.406 1.024

PeF (cm) 50 0 50 1.381 3.154 2.166 0.411

219

Table 67 Correlation matrix (Pearson (n)) for R. rita

Variables Wt FL TL HL St DF CF AF AdF PF PeF

Wt 1 1.000 1.000 0.999 1.000 1.000 1.000 0.999 0.990 1.000 0.998

FL 1.000 1 1.000 0.999 1.000 1.000 1.000 0.999 0.990 1.000 0.998

TL 1.000 1.000 1 0.999 1.000 1.000 1.000 0.999 0.990 1.000 0.998

HL 0.999 0.999 0.999 1 0.999 0.999 0.999 0.998 0.991 0.999 0.997

St 1.000 1.000 1.000 0.999 1 1.000 1.000 0.999 0.990 1.000 0.998

DF 1.000 1.000 1.000 0.999 1.000 1 0.999 0.998 0.990 0.999 0.998

CF 1.000 1.000 1.000 0.999 1.000 0.999 1 0.998 0.989 1.000 0.998

AF 0.999 0.999 0.999 0.998 0.999 0.998 0.998 1 0.987 0.998 0.997

AdF 0.990 0.990 0.990 0.991 0.990 0.990 0.989 0.987 1 0.989 0.990

PF 1.000 1.000 1.000 0.999 1.000 0.999 1.000 0.998 0.989 1 0.998

PeF 0.998 0.998 0.998 0.997 0.998 0.998 0.998 0.997 0.990 0.998 1

Table 68 Bartlett's sphericity test for R. rita

Chi-square (Observed value) -Inf

220

Chi-square (Critical value) 73.311

DF 55 p-value -

Alpha 0.05

H0 There is no correlation significantly different from 0 between the variables.

Ha At least one of the correlations between the variables is significantly different from 0.

Table 69 Eigen values of Principal Component Analysis f (PCA) for R. rita

F1 F2 F3 F4 F5 F6 F7 F8

Eigenvalue 10.654 0.238 0.092 0.015 0.000 0.000 0.000 0.000

Variability (%) 96.852 2.164 0.840 0.139 0.003 0.001 0.001 0.001

Cumulative % 96.852 99.016 99.856 99.994 99.997 99.998 99.999 100.000

Figure 15 Scree plot between Eigen values, factors and cumulative variability for R. rita

221

222

Table 70 Eigen vectors of PCA for R. rita

F1 F2 F3 F4 F5 F6 F7 F8

Wt 0.306 -0.085 -0.097 -0.121 -0.090 -0.055 -0.005 -0.338

FL 0.306 -0.085 -0.097 -0.121 -0.090 -0.055 -0.005 -0.338

TL 0.306 -0.086 -0.098 -0.130 -0.199 -0.530 -0.531 0.519

HL 0.306 -0.087 -0.099 -0.132 0.924 0.049 -0.035 0.122

St 0.306 -0.085 -0.097 -0.121 -0.090 -0.055 -0.005 -0.338

DF 0.304 -0.110 -0.143 0.936 0.006 0.001 -0.005 0.008

CF 0.306 -0.068 -0.099 -0.120 -0.153 -0.121 0.817 0.415

AF 0.293 -0.078 0.952 0.041 0.005 -0.015 0.005 -0.003

AdF 0.271 0.962 -0.005 0.024 0.003 0.003 -0.015 -0.005

PF 0.306 -0.087 -0.080 -0.125 -0.227 0.831 -0.219 0.295

PeF 0.306 -0.085 -0.097 -0.121 -0.090 -0.055 -0.005 -0.338

Table 71 Factor loadings of PCA for R. rita

F1 F2 F3 F4 F5 F6 F7 F8 Wt 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

FL 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

223

TL 0.998 -0.042 -0.030 -0.016 -0.003 -0.007 -0.006 0.004

HL 0.998 -0.042 -0.030 -0.016 0.016 0.001 0.000 0.001

St 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

DF 0.991 -0.053 -0.043 0.116 0.000 0.000 0.000 0.000

CF 0.999 -0.033 -0.030 -0.015 -0.003 -0.001 0.009 0.003

AF 0.956 -0.038 0.289 0.005 0.000 0.000 0.000 0.000

AdF 0.883 0.469 -0.002 0.003 0.000 0.000 0.000 0.000

PF 0.999 -0.042 -0.024 -0.015 -0.004 0.010 -0.002 0.002

PeF 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

Table 72 Correlations between variables and factors of PCA for R. rita

F1 F2 F3 F4 F5 F6 F7 F8

Wt 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

FL 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

TL 0.998 -0.042 -0.030 -0.016 -0.003 -0.007 -0.006 0.004

HL 0.998 -0.042 -0.030 -0.016 0.016 0.001 0.000 0.001

St 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

224

DF 0.991 -0.053 -0.043 0.116 0.000 0.000 0.000 0.000

CF 0.999 -0.033 -0.030 -0.015 -0.003 -0.001 0.009 0.003

AF 0.956 -0.038 0.289 0.005 0.000 0.000 0.000 0.000

AdF 0.883 0.469 -0.002 0.003 0.000 0.000 0.000 0.000

PF 0.999 -0.042 -0.024 -0.015 -0.004 0.010 -0.002 0.002

PeF 0.999 -0.041 -0.030 -0.015 -0.002 -0.001 0.000 -0.003

Figure 16 Graph between variables for R. rita

225

Table 73 Contribution of the variables (%) of PCA for R. rita

F1 F2 F3 F4 F5 F6 F7 F8

Wt 9.360 0.722 0.946 1.452 0.805 0.299 0.002 11.414

FL 9.360 0.722 0.946 1.452 0.805 0.299 0.002 11.414

TL 9.358 0.745 0.968 1.684 3.972 28.069 28.249 26.956

HL 9.356 0.749 0.984 1.747 85.322 0.237 0.124 1.482

St 9.360 0.722 0.946 1.452 0.805 0.299 0.002 11.414

DF 9.217 1.200 2.035 87.535 0.004 0.000 0.003 0.006

CF 9.365 0.458 0.981 1.441 2.328 1.457 66.786 17.183

AF 8.587 0.612 90.606 0.167 0.003 0.022 0.003 0.001

AdF 7.318 92.595 0.003 0.058 0.001 0.001 0.022 0.003

PF 9.361 0.754 0.638 1.559 5.150 69.019 4.804 8.714

PeF 9.360 0.722 0.946 1.452 0.805 0.299 0.002 11.414

Table 74 Squared cosines of the variables of PCA for R. rita

F1 F2 F3 F4 F5 F6 F7 F8

Wt 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

226

FL 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

TL 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

HL 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

St 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

DF 0.982 0.003 0.002 0.013 0.000 0.000 0.000 0.000

CF 0.998 0.001 0.001 0.000 0.000 0.000 0.000 0.000

AF 0.915 0.001 0.084 0.000 0.000 0.000 0.000 0.000

AdF 0.780 0.220 0.000 0.000 0.000 0.000 0.000 0.000

PF 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000

PeF 0.997 0.002 0.001 0.000 0.000 0.000 0.000 0.000 Figure 17 Graph between Observations for R. rita

227

Figure 18 Biplot graph between factors for R. rita

228

D Principle Component Analysis (PCA) of Sperata seenghala

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among the S. seenghala collected samples from different geographical locations along with differentiation between and within the groups.

The summary statistics, correlation matrix (Pearson (n)) of PCA and Bartlett’s sphericity test of PCA are shown in Table 75, Table 76 and Table 77, respectively. The results obtained from the PCA indicated clearly that the augment in the

229

number of factors or components was correlated with the decrease in Eigen values (Figure 19). The values showed that its trend reached its maximum at level of second factor (Table 78). The first two main factors accounted for 99.907% of cumulative variability. Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 99.91% of the cumulative variation among the morphometric parameters of study (Figure 20). The first group amongst the major two groups accounted for 97.154% of the cumulative variability while the second accounted for 2.752% of the cumulative variability. The factor one (F1) was positively correlated when variables of morphometric characters were compared with factors. In case of factor F2 and F3, only the average length of the paired pelvic fins and average length of pectoral fins & the average length of the paired pelvic fins, respectively, were positively correlated, others were negatively correlated. In case of F4, the values of Head length, stoutness, the average length of the paired pelvic fins and average length of paired pectoral fins were positively correlated. In case of F5, the values of head length, anal fin length, the average length of the paired pelvic fins and average length of pectoral fins were positively correlated. The others were negative in their relation. The relationship was positive in case of factor six (F6) when parameters like stoutness, anal fin length, average length of paired pectoral fins and the average length of the paired pelvic fins were studied. In F7 head length, stoutness, caudal fin, anal fin length, average length of pectoral fins and the average length of the paired pelvic fins were positive while other parameters were negatively correlated. Total length was positive in factor F 8 in addition to positive parameters in factor F7. In the F9 only the body weight and dorsal fin length were negative in correlation, while the others were positively correlated (Table 81).

The observation plot and bi-plot of the parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups. On the observation of the these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on observation plot analysis (Figure 21) and the F1 & F2 factors bi-plot (Figure 22). The Eigen vectors, factor loading values and the percentage

230

contribution of the factors are presented in Tables 79, 80 and 82, respectively. The bold values in the squared cosines of the variables showed, which are the major two factors on the basis of which the parameters were studied (Table 83).

Table 75 Summary statistics for S. seenghala

Variable Obs Obs. Obs. Minimum Maximum Mean Std. with without deviation missing missing data data Wt (g) 50 0 50 1200.000 3000.000 1881.280 465.373

FL (cm) 50 0 50 23.774 59.436 37.276 9.220

TL (cm) 50 0 50 29.566 73.914 46.363 11.471

HL (cm) 50 0 50 6.706 16.764 10.537 2.574

St (cm) 50 0 50 12.497 31.242 19.622 4.853

DF (cm) 50 0 50 3.981 9.952 6.241 1.544

CF (cm) 50 0 50 5.791 14.478 9.083 2.246

AF (cm) 50 0 50 2.438 6.096 3.827 0.943 AdF 50 0 50 1.524 3.810 2.389 0.591 (cm) PF (cm) 50 0 50 2.743 6.858 4.313 1.064

231

PeF 50 0 50 2.438 6.096 3.669 0.973 (cm)

Table 76 Correlation matrix (Pearson (n)) for S. seenghala

Variables Wt FL TL HL St DF CF AF AdF PF PeF Wt 1 1.000 1.000 0.999 0.999 1.000 1.000 1.000 1.000 0.997 0.822 FL 1.000 1 1.000 0.999 0.999 1.000 1.000 1.000 1.000 0.997 0.822 TL 1.000 1.000 1 0.998 0.999 1.000 1.000 1.000 1.000 0.997 0.822 HL 0.999 0.999 0.998 1 0.999 0.999 0.998 0.998 0.999 0.995 0.821 St 0.999 0.999 0.999 0.999 1 0.999 0.999 0.999 0.999 0.996 0.823 DF 1.000 1.000 1.000 0.999 0.999 1 1.000 1.000 1.000 0.997 0.822 CF 1.000 1.000 1.000 0.998 0.999 1.000 1 0.999 1.000 0.997 0.822 AF 1.000 1.000 1.000 0.998 0.999 1.000 0.999 1 1.000 0.996 0.822 AdF 1.000 1.000 1.000 0.999 0.999 1.000 1.000 1.000 1 0.997 0.822 PF 0.997 0.997 0.997 0.995 0.996 0.997 0.997 0.996 0.997 1 0.821 PeF 0.822 0.822 0.822 0.821 0.823 0.822 0.822 0.822 0.822 0.821 1

Table 77 Bartlett's sphericity test for S. seenghala

Chi-square (Observed value) -Inf

Chi-square (Critical value) 73.311

232

DF 55 p-value - alpha 0.05

H0 There is no correlation significantly different from 0 between the variables.

Ha At least one of the correlations between the variables is significantly different from 0.

Table 78 Eigen values of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9 10.687 0.303 0.006 0.003 0.001 0.001 0.000 0.000 0.000 Eigenvalue 97.154 2.752 0.054 0.024 0.007 0.006 0.001 0.000 0.000 Variability (%) 97.154 99.907 99.960 99.985 99.992 99.998 100.000 100.000 100.000 Cumulative % Figure 19 Scree plot between Eigen values, factors and cumulative variability for S. seenghala

233

Table 79 Eigen vectors of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9

Wt 0.306 -0.083 -0.076 -0.147 -0.060 -0.167 -0.162 -0.228 -0.310

FL 0.306 -0.082 -0.077 -0.149 -0.061 -0.173 -0.181 -0.312 0.841

TL 0.306 -0.082 -0.078 -0.152 -0.062 -0.191 -0.303 0.858 0.056

HL 0.305 -0.084 -0.233 0.864 0.271 -0.158 0.005 0.001 0.001

St 0.305 -0.079 -0.142 0.166 -0.625 0.679 0.030 0.017 0.004

234

DF 0.306 -0.083 -0.076 -0.147 -0.060 -0.167 -0.162 -0.228 -0.310

CF 0.306 -0.082 -0.080 -0.162 -0.076 -0.237 0.891 0.102 0.020

AF 0.305 -0.082 -0.096 -0.265 0.716 0.554 0.038 0.016 0.005

AdF 0.306 -0.083 -0.076 -0.147 -0.060 -0.167 -0.162 -0.228 -0.310

PF 0.305 -0.080 0.938 0.138 0.015 0.029 0.007 0.004 0.001

PeF 0.259 0.966 -0.002 0.001 0.002 -0.002 -0.001 -0.001 0.000

Table 80 Factor loadings of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9

Wt 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

FL 0.999 -0.045 -0.006 -0.008 -0.002 -0.004 -0.002 -0.002 0.002

TL 0.999 -0.045 -0.006 -0.008 -0.002 -0.005 -0.004 0.006 0.000

HL 0.998 -0.046 -0.018 0.045 0.008 -0.004 0.000 0.000 0.000

St 0.999 -0.044 -0.011 0.009 -0.018 0.018 0.000 0.000 0.000

DF 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

CF 0.999 -0.045 -0.006 -0.008 -0.002 -0.006 0.010 0.001 0.000

AF 0.999 -0.045 -0.007 -0.014 0.020 0.014 0.000 0.000 0.000

235

AdF 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

PF 0.996 -0.044 0.072 0.007 0.000 0.001 0.000 0.000 0.000

PeF 0.847 0.531 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Table 81 Correlations between variables and factors of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9

Wt 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

FL 0.999 -0.045 -0.006 -0.008 -0.002 -0.004 -0.002 -0.002 0.002

TL 0.999 -0.045 -0.006 -0.008 -0.002 -0.005 -0.004 0.006 0.000

HL 0.998 -0.046 -0.018 0.045 0.008 -0.004 0.000 0.000 0.000

St 0.999 -0.044 -0.011 0.009 -0.018 0.018 0.000 0.000 0.000

DF 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

CF 0.999 -0.045 -0.006 -0.008 -0.002 -0.006 0.010 0.001 0.000

AF 0.999 -0.045 -0.007 -0.014 0.020 0.014 0.000 0.000 0.000

AdF 0.999 -0.046 -0.006 -0.008 -0.002 -0.004 -0.002 -0.001 -0.001

PF 0.996 -0.044 0.072 0.007 0.000 0.001 0.000 0.000 0.000

PeF 0.847 0.531 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Figure 20 Graph between variables for S. seenghala

236

Table 82 Contribution of the variables (%) of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9

Wt 9.337 0.684 0.579 2.152 0.355 2.788 2.628 5.201 9.610

FL 9.337 0.680 0.589 2.216 0.376 2.981 3.272 9.743 70.807

TL 9.337 0.665 0.611 2.308 0.381 3.660 9.165 73.555 0.317

HL 9.315 0.704 5.433 74.702 7.343 2.500 0.002 0.000 0.000

St 9.332 0.632 2.008 2.760 39.029 46.117 0.092 0.029 0.002

237

DF 9.337 0.684 0.579 2.152 0.355 2.788 2.628 5.201 9.610

CF 9.336 0.668 0.648 2.624 0.583 5.614 79.440 1.046 0.042

AF 9.330 0.679 0.923 7.023 51.199 30.679 0.141 0.024 0.003

AdF 9.337 0.684 0.579 2.152 0.355 2.788 2.628 5.201 9.610

PF 9.290 0.637 88.050 1.911 0.023 0.084 0.004 0.001 0.000

PeF 6.714 93.284 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Table 83 Squared cosines of the variables of PCA for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9

Wt 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

FL 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

TL 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

HL 0.995 0.002 0.000 0.002 0.000 0.000 0.000 0.000 0.000

St 0.997 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

DF 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

238

CF 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

AF 0.997 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

AdF 0.998 0.002 0.000 0.000 0.000 0.000 0.000 0.000 0.000

PF 0.993 0.002 0.005 0.000 0.000 0.000 0.000 0.000 0.000

PeF 0.718 0.282 0.000 0.000 0.000 0.000 0.000 0.000 0.000 Values in bold correspond for each variable to the factor for which the squared cosine is the largest

Figure 21 Graph between Observations for S. seenghala

239

Figure 22 Biplot graph between factors for S. seenghala

E Principle Component Analysis (PCA) of Wallago attu

The Pearson type Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 for further clarification of the genetic relationships among the W. attu collected samples from different geographical locations along with differentiation between and within the groups.

The summary statistics, correlation matrix (Pearson (n)) of PCA and Bartlett’s sphericity test of PCA are shown in Table 84, Table 85 and Table 86, respectively. The results obtained from the PCA indicated clearly that the augment in the number of factors or components was correlated with the decrease in Eigen values (Figure 23). The values showed that

240

its trend reached its maximum at level of second factor (Table 90). In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first two main factors accounted for 99.996% of cumulative variability. Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 100% of the cumulative variation among the morphometric parameters of study (Figure 24). The first group amongst the major two groups accounted for 99.983% of the cumulative variability while the second accounted for 0.013% of the cumulative variability. The factor one (F1) was positively correlated with all the body parameters. Correlation was positive for head length only for F2 and head length and stoutness for F3, while others were negatively correlated. In factor four (F4) only the total length was negatively correlated and in factor five (F5) fork length, total length, head length and stoutness were positive in correlation (Table 89).

The graphical presentation shows the correlation amongst the loadings and factors by PCA (Figure 14). The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups (Figures 25 and 26). On the observation of the these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 factors bi-plot (Figure 26) and observation plot analysis (Figure 25). The Eigen vectors, factor loading values and the percentage contribution of the factors are presented in Tables 87, 88 and 91, respectively. The bold values in the squared cosines of the variables showed, which are the major two factors on the basis of which the parameters were studied (Table 92).

Table 84 Summary statistics for W. attu

with missing without Variable Obs Minimum Maximum Mean Std. deviation data missing data

241

Wt (g) 50 0 50 1155.000 2980.000 2002.180 566.048

FL (cm) 50 0 50 26.697 68.880 46.298 13.070

TL (cm) 50 0 50 29.044 74.935 50.367 14.223

HL (cm) 50 0 50 6.454 16.652 11.178 3.164

St (cm) 50 0 50 10.561 27.249 18.320 5.159

DF (cm) 50 0 50 3.520 9.083 6.103 1.725

CF (cm) 50 0 50 2.640 6.812 4.577 1.294

PF (cm) 50 0 50 2.347 6.055 4.068 1.150

PeF (cm) 50 0 50 3.520 9.083 6.103 1.725

Table 85 Correlation matrix (Pearson (n)) for W. attu

Variables Wt FL TL HL St DF CF PF PeF

Wt 1 1.000 1.000 0.999 1.000 1.000 1.000 1.000 1.000

FL 1.000 1 1.000 0.999 1.000 1.000 1.000 1.000 1.000

TL 1.000 1.000 1 0.999 1.000 1.000 1.000 1.000 1.000

HL 0.999 0.999 0.999 1 0.999 0.999 0.999 0.999 0.999

St 1.000 1.000 1.000 0.999 1 1.000 1.000 1.000 1.000

DF 1.000 1.000 1.000 0.999 1.000 1 1.000 1.000 1.000

242

CF 1.000 1.000 1.000 0.999 1.000 1.000 1 1.000 1.000

PF 1.000 1.000 1.000 0.999 1.000 1.000 1.000 1 1.000

PeF 1.000 1.000 1.000 0.999 1.000 1.000 1.000 1.000 1

Table 86 Bartlett's sphericity test for W. attu

Chi-square (Observed value) -Inf Chi-square (Critical value) 61.656 DF 45 p-value - alpha 0.05

H0 There is no correlation significantly different from 0 between the variables. Ha At least one of the correlations between the variables is significantly different from 0.

Table 87 Eigen values of PCA for W. attu

F1 F2 F3 F4 F5 9.998 0.001 0.000 0.000 0.000 Eigenvalue 99.983 0.013 0.002 0.001 0.001 Variability (%) 99.983 99.996 99.998 99.999 100.000 Cumulative % Figure 23 Scree plot between Eigen values, factors and cumulative variability for W. attu

243

Table 88 Eigen vectors of PCA for W. attu

F1 F2 F3 F4 F5

Wt 0.316 -0.102 -0.089 -0.045 -0.215

FL 0.316 -0.108 -0.223 0.802 0.442

TL 0.316 -0.106 -0.187 -0.585 0.715

HL 0.316 0.949 0.015 0.003 0.008

St 0.316 -0.122 0.931 0.049 0.126

244

DF 0.316 -0.102 -0.089 -0.045 -0.215

CF 0.316 -0.102 -0.089 -0.045 -0.215

PF 0.316 -0.102 -0.089 -0.045 -0.215

PeF 0.316 -0.102 -0.089 -0.045 -0.215

Table 89 Factor loadings of PCA for W. attu

F1 F2 F3 F4 F5

Wt 1.000 -0.004 -0.001 0.000 -0.002

FL 1.000 -0.004 -0.003 0.008 0.004

TL 1.000 -0.004 -0.003 -0.006 0.006

HL 0.999 0.034 0.000 0.000 0.000

St 1.000 -0.004 0.014 0.001 0.001

DF 1.000 -0.004 -0.001 0.000 -0.002

245

CF 1.000 -0.004 -0.001 0.000 -0.002

PF 1.000 -0.004 -0.001 0.000 -0.002

PeF 1.000 -0.004 -0.001 0.000 -0.002

Table 90 Correlations between variables and factors of PCA for W. attu

F1 F2 F3 F4 F5

Wt 1.000 -0.004 -0.001 0.000 -0.002

FL 1.000 -0.004 -0.003 0.008 0.004

TL 1.000 -0.004 -0.003 -0.006 0.006

HL 0.999 0.034 0.000 0.000 0.000

St 1.000 -0.004 0.014 0.001 0.001

DF 1.000 -0.004 -0.001 0.000 -0.002

CF 1.000 -0.004 -0.001 0.000 -0.002

PF 1.000 -0.004 -0.001 0.000 -0.002

PeF 1.000 -0.004 -0.001 0.000 -0.002

246

Figure 24 Graph between variables for W. attu

Table 91 Contribution of the variables (%) of PCA for W. attu

F1 F2 F3 F4 F5

Wt 10.001 1.042 0.798 0.200 4.625

FL 10.001 1.160 4.993 64.300 19.547

247

TL 10.001 1.132 3.489 34.255 51.123

HL 9.990 89.980 0.023 0.001 0.006

St 9.999 1.478 86.704 0.242 1.577

DF 10.001 1.042 0.798 0.200 4.625

CF 10.001 1.042 0.798 0.200 4.625

PF 10.001 1.042 0.798 0.200 4.625

PeF 10.001 1.042 0.798 0.200 4.625

Table 92 Squared cosines of the variables of PCA for W. attu

F1 F2 F3 F4 F5

Wt 1.000 0.000 0.000 0.000 0.000

FL 1.000 0.000 0.000 0.000 0.000

TL 1.000 0.000 0.000 0.000 0.000

HL 0.999 0.001 0.000 0.000 0.000

St 1.000 0.000 0.000 0.000 0.000

248

DF 1.000 0.000 0.000 0.000 0.000

CF 1.000 0.000 0.000 0.000 0.000

PF 1.000 0.000 0.000 0.000 0.000

PeF 1.000 0.000 0.000 0.000 0.000

Values in bold correspond for each variable to the factor for which the squared cosine is the largest

Figure 25 Graph between Observations for W. attu

249

Figure 26 Biplot graph between factors for W. attu

4.3 Measurement of the Physicochemical Parameters

The physico-chemical parameters of water which were analyzed from the samples collected from the sampling sites and their averages were calculated. These parameters guesstimate the water quality and are interspersed and sway the biological productivity of the water and growth of the flora and fauna in the Riverine as well as ponds waters are discussed as under.

Table 93 Average Physicochemical Parameters of the Study Sites

250

Site Temp. pH D.O. E.C. Sal. T.D.S. T.A. T.H.

Chashma Barrage 26.860 7.86 7.95 1.100 0.68 1289.00 134.45 239.00 Trimu Barrage 24.66 8.62 7.84 0.450 0.45 1060.67 145.87 80.45 Taunsa Barrage 25.27 8.45 7.18 0.386 0.48 955.44 155.00 73.63 Qadirabad Barrage 24.23 8.54 8.45 0.412 0.47 1077.56 122.45 87.45 Baloki Barrage 25.27 7.95 6.49 1.323 0.77 1583.21 247.22 122.23 Temp. = Water Temperature (oC), DO=Dissolved Oxygen (ppm), E.C. = Electrical Conductivity (mScm-1, Sal. = Salinity (ppt), TDS =Total Dissolved Solids (mgL-1), T.A. =Total Alkalinity (mgL-1), T.H. =Total Hardness (mgL-1)

A. Water Temperature The water temperature was noted from three different locations of the sampling sites and their averages were calculated which remained as 26.86, 24.66, 25.27, 24.23 and 25.27°C for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93). B. pH The average pH remained as 7.86, 8.62, 8.45, 8.54 and 7.95 for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

C. Dissolved Oxygen (DO) The average DO values remained as 7.95, 7.84, 7.18, 8.45 and 6.49 ppm for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

251

D. Electrical Conductivity (EC) The average EC values remained as 1.100, 0.450, 0.386, 0.412 and 1.323 mScm-1 for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

E. Salinity Average salinity values for different sites remained as 0.68, 0.48, 0.47, 0.45 and 0.77 ppt for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

F. Total Dissolved Solids (TDS) The total dissolved solids average values for different sites remained as 1349, 1060.67, 955.44, 1077.56 and 1583.21 mgL-1 for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

G. Total Alkalinity The total alkalinity average values for different sited remained as 134.45, 145.87, 155.00, 122.45 and 247.22 mgL-1 for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93).

H. Total Hardness Total hardness average values for different sited remained as 239.00, 80.45, 73.63, 87.45and 122.23 mgL-1 for the Chashma barrage, Taunsa from River Indus, Trimu from the junction of River Chenab and Jhelum, Qadirabad at River Chenab and Baloki at River Ravi from where the samples were collected, respectively (Table 93)

252

Table 94 Correlation Matrix of Physico-chemical Parameters of the Study Sites

Temp. pH D.O. E.C. Sal. T.D.S. T.A. 0.107 pH 0.864 0.313 0.905 D.O. 0.608 0.034 0.482 -0.798 -0.669 E.C. 0.411 0.106 0.217 0.193 -0.888 -0.828 0.925 Sal. 0.756 0.044 0.083 0.025 0.172 -0.857 -0.809 0.889 0.994 T.D.S. 0.782 0.064 0.097 0.044 0.001 -0.484 -0.736 -0.930 0.452 0.717 0.734 T.A. 0.409 0.156 0.022 0.445 0.173 0.158 0.801 -0.499 -0.300 0.906 0.699 0.657 0.048 T.H. 0.103 0.392 0.624 0.034 0.189 0.229 0.938 Cell Contents Pearson Correlation P-value Bold values are significantly different

Correlation matrix of physicochemical parameters of the study sites

253

The data obtained from the present experimental sites was subjected to statistical analysis to find out the correlation coefficient among different physic-chemical parameters. The correlation matrix of various physiochemical characteristics of the study sites, showed the presence of different patterns of correlation with one another (Table 94).

The correlation of the pH with water temperature (r = 0.107) and dissolved oxygen (r = 0.905) was positively non- significant while the correlation with electrical conductivity (r = -0.798), salinity (r = -0.888), total dissolved solids (r = - 0.857), total alkalinity (r = -0.736) and total hardness (r = -0.499) was negatively non-significant. The correlation of the dissolved oxygen with water temperature (r = 0.313) was positively non-significant while the correlation with electrical conductivity (r = -0.669), salinity (r = -0.828), total dissolved solids (r = -0.809), total alkalinity (r = -0.930) and total hardness (r = -0.300) was negative but also non-significant as like with the water temperature. The electrical conductivity was positively correlated with all the physic-chemical parameters as with water temperature (r = 0.482), salinity (r = 0.925), total dissolved solids (r = 0.889), total alkalinity (r = 0.452) and total hardness (r = 0.906) and this correlation was non-significant.

The salinity amongst the water parameters was correlated positively with water temperature (r = 193), total alkalinity (r = 0.717) and total hardness (r = 0.734) and it was non-significant but with total dissolved solids (r = 0.994) the correlation was also positive but highly significant (P <0.001).The total dissolved solids values observed from the study sites were positively correlated with water temperature (r = 0.172), total alkalinity (r = 0.734) and total harness (r = 0.657) and this correlation was non-significant. The correlation between the total alkalinity and total hardness was also positive and non- significant (r = 0.048).

4.4 DNA Extraction

254

Total genomic DNA isolation was carried out from the stored fish samples using the procedure described by Lopera-Barrero et al., (2008). This procedure is given in detail in the chapter 3.

A Primers used for Amplification

a Channa punctatus The DNA collected from fishes collected from different locations was amplified with the primers given in Table 95. The products after Polymerase Chain Reactions (PCR) were subjected to the gel electrophoresis for measuring qualitatively with Gel documentation. The data from all the scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. Percent Polymorphism found from the data obtained from the samples of Channa punctatus collected from all study sites is also shown in Table 95. The number of bands for PCR products ranged from as low as three to a maximum of seven, with an average of 6 bands per primer. The number of polymorphic bands per primer was 1 to 3 amplified. The polymorphic bands in these populations ranged from 14.29% to 50%.

Table 95 Primers used and amplification data for Channa punctatus

Sr. Name of Primer Sequence of the Number of total Number of total Number of total % Polymorphism No. primer amplified monomorphic polymorphic bands bands bands

1 OPB-02 TGATCCCTGG 6 3 3 50

2 OPB-06 TGCTCTGCCC 6 3 3 50

3 OPC-11 AAAGCTGCGG 5 4 1 20

255

4 OPC-13 AAGCCTCGTC 7 6 1 14.29

5 OPC-15 GACGGATCAG 7 6 1 14.29

6 OPD-01 ACCGCGAAGG 5 4 1 20

7 OPD-02 GGACCCAACC 5 3 2 40 8 OPD-03 GTCGCCGTCA 5 3 2 40

9 OPD-04 TCTGGTGAGG 7 4 3 42.86

10 OPD-05 TGAGCGGACA 4 3 1 25

b Channa marulius

The DNA collected from fishes collected from different locations was amplified with the primers given in Table 96. The products after Polymerase Chain Reactions (PCR) were subjected to the gel electrophoresis for measuring qualitatively with Gel documentation. All the scorable amplified bands data was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. The Table 96 also shows the % Polymorphism found in the samples of Channa marulius collected from the all study sites. The bands of amplification products produced by primers in Channa marulius ranged from as low as three to a maximum of seven, with an average of 6 bands per primer. 1 to 3 polymorphic bands per primer were obtained. The polymorphic bands in these populations ranged from 14.29% to 50%.

Table 96 Primers used and amplification data for Channa marulius

256

Sr. Name of Sequence of the primer Number of Number of total Number of total % Polymorphism No. Primer total amplified monomorphic bands polymorphic bands bands

1 OPB-02 TGATCCCTGG 6 3 3 50.00

2 OPB-06 TGCTCTGCCC 6 3 3 50.00

3 OPC-11 AAAGCTGCGG 5 3 2 40.00

4 OPC-13 AAGCCTCGTC 7 6 1 14.29

5 OPC-15 GACGGATCAG 7 6 1 14.29

6 OPD-01 ACCGCGAAGG 5 4 1 20.00

7 OPD-02 GGACCCAACC 5 4 1 20.00

8 OPD-03 GTCGCCGTCA 5 3 2 40.00

9 OPD-04 TCTGGTGAGG 7 5 2 28.57

10 OPD-05 TGAGCGGACA 3 2 1 33.33

c Rita rita

The DNA collected from fishes collected from different locations was amplified with the primers given in Table 97. The products after Polymerase Chain Reactions (PCR) were subjected to the gel electrophoresis for measuring qualitatively with Gel documentation. All the scorable amplified bands data was organized on the one-zero pattern i.e. the presence of 257

band recorded as “1” and for absence marked as “0”. The Table 97 also shows the % Polymorphism found in the samples of Rita rita collected from the all study sites. The number of bands in the PCR product produced by primers in Rita rita ranged from as low as three to a maximum of seven, with an average of 5.5 bands per primer. 1 to 3 polymorphic bands per primer were observed. The polymorphic bands in these populations ranged from 14.29% to 50%.

Table 97 Primers used and amplification data for Rita rita

Sr. No. Name of Sequence of the primer Number of Number of total Number of total % Polymorphism Primer total monomorphic bands polymorphic bands amplified bands

1 OPB-02 TGATCCCTGG 6 3 3 50.00

2 OPB-06 TGCTCTGCCC 6 4 2 33.33

3 OPC-11 AAAGCTGCGG 5 4 1 20.00

4 OPC-13 AAGCCTCGTC 7 6 1 14.29

5 OPC-15 GACGGATCAG 7 6 1 14.29

6 OPD-01 ACCGCGAAGG 5 4 1 20.00

7 OPD-02 GGACCCAACC 5 4 1 20.00

8 OPD-03 GTCGCCGTCA 5 3 2 40.00

9 OPD-04 TCTGGTGAGG 6 4 2 33.33

10 OPD-05 TGAGCGGACA 3 2 1 33.33

258

d Sperata seenghala

The DNA collected from fishes collected from different locations was amplified with the primers given in Table 98. The products after Polymerase Chain Reactions (PCR) were subjected to the gel electrophoresis for measuring qualitatively with Gel documentation. All the scorable amplified bands data was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. The Table 98 also shows the % Polymorphism found in the samples of S. seenghala collected from the all study sites. The bands produced in amplified products by primers in PCR ranged from as low as three to a maximum of seven, with an average of 5.5 bands per primer in S. seenghala. 1 to 4 polymorphic bands per primer were found. The polymorphic bands in these populations ranged from 14.29% to 57.14%.

Table 98 Primers used and amplification data for Sperata seenghala

Sr. No. Name of Sequence of the primer Number of Number of total Number of total % Polymorphism Primer total monomorphic bands polymorphic bands amplified bands

1 OPB-02 TGATCCCTGG 6 5 1 16.67

2 OPB-06 TGCTCTGCCC 5 3 2 40.00

3 OPC-11 AAAGCTGCGG 6 4 2 33.33

4 OPC-13 AAGCCTCGTC 5 4 1 20.00

259

5 OPC-15 GACGGATCAG 7 6 1 14.29

6 OPD-01 ACCGCGAAGG 5 4 1 20.00

7 OPD-02 GGACCCAACC 4 2 2 50.00

8 OPD-03 GTCGCCGTCA 7 5 2 28.57

9 OPD-04 TCTGGTGAGG 7 3 4 57.14

10 OPD-05 TGAGCGGACA 3 2 1 33.33

e Wallago attu

The DNA collected from fishes collected from different locations was amplified with the primers given in Table 99. The products after Polymerase Chain Reactions (PCR) were subjected to the gel electrophoresis for measuring qualitatively with Gel documentation. All the scorable amplified bands data was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. The Table 99 also shows the % Polymorphism found in the samples of Wallago attu collected from the all study sites. The number bands ranged in W. attu from as low as three to a maximum of seven, with an average of 5.5 bands per primer in the amplified products. 1 to 4 polymorphic bands per primer were observed. The polymorphic bands in these populations ranged from 14.29% to 66.67%.

Table 99 Primers used and amplification data for Wallago attu

260

Sr. No. Name of Sequence of the primer Number of Number of total Number of total % Polymorphism Primer total monomorphic bands polymorphic bands amplified bands

1 OPB-02 TGATCCCTGG 5 2 3 60.00

2 OPB-06 TGCTCTGCCC 6 2 4 66.67

3 OPC-11 AAAGCTGCGG 6 5 1 16.67

4 OPC-13 AAGCCTCGTC 5 4 1 20.00

5 OPC-15 GACGGATCAG 7 6 1 14.29

6 OPD-01 ACCGCGAAGG 5 4 1 20.00

7 OPD-02 GGACCCAACC 4 3 1 25.00

8 OPD-03 GTCGCCGTCA 7 6 1 14.29

9 OPD-04 TCTGGTGAGG 7 2 5 71.43

10 OPD-05 TGAGCGGACA 3 1 2 66.67

4.5 Quantification of DNA

261

The PCR product of the DNA samples were loaded into the gel after mixing with 10X DNA loading buffer with 0.21% bromophenol blue, 0.21% xylene cyanol FF, 0.2 molar EDTA and 50% glycerol. The selected representative pictures of the gel electrophoresis are given in Figures 27 to 34. Note - The band size of the used marker in Figures; 27, 31, 33 and 34 is same as given in Figure 27, while the used marker in Figures 28, 29, 30 and 32 is same as in 32.

1 2 3 4 5 6 7 8 M

1000bp

800bp

500bp

260bp

Figure 27 Picture showing the Amplification of the OPB-2 for R. rita (1-4) and S. seenghala (5-8) for the samples collected from Baloki

262

1 2 3 4 5 6 7 M

Figure 28 Picture showing the Amplification of OPD-4 for W. attu for the samples collected from Baloki

263

1 2 3 4 5 6 7 8 M

Figure 29 Picture showing the Amplification of OPD-4 for R. rita (1-4) and S. seenghala (5-8) for the samples collected from Qadirabad

1 2 3 4 5 6 7 M

264

Figure 30 Picture showing the Amplification of OPC-11 for C. punctatus (1-3) and C. marulius (4-7) for the samples collected from Baloki, River Ravi.

265

1 2 3 4 5 6 7 8 M

Figure 31 Picture showing the Amplification of OPD-5 for C. marulius (1-4) and C. punctatus (5-8) for collected samples from Trimu and Taunsa barrages

1 2 3 4 5 6 7 8 9 10 M

266

1000bp

500b p 300bp

200bp

Figure 32 Picture showing the Amplification of OPD-1 for S. seenghala (1-5) and OPD-4 for R. rita (6-10) from Taunsa barrage.

1 2 3 4 5 6 7 8 M 9 10 11 12 13

267

Figure 33 Picture showing the Amplification of OPB-2 for W. attu for samples collected from Trimu and Taunsa Barrages.

1 2 3 4 5 6 7 8 9 10

268

Figure 34 Picture showing the amplification of OPC-11 and OPC-15 for S. seenghala for samples from Qadirabad and Chashma barrages

4.6 Agglomerative Hierarchical Clustering (AHC) from PCR Products/Agarose Gel Electrophoresis of Experimental Species

The data obtained from the random markers for the Randomly Amplified Polymorphic DNA (RAPD) was subjected to statistical analysis for Jaccard’s coefficient (JACCORD 1901) by following the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) for Hierarchical Clustering of the similar groups on the basis of similarity amongst the genotypes and the dendrogram generation, by XLSTAT 2012 version 1.02, computer software. The Principal Component Analysis (PCA) for grouping of the different genotypes from the different environmental conditions was done by Spearman

269

Varimax rotation method for bi-plot generation of the co-occurrence of the same genotypes with similar genetic properties and specificity of different primers in the same program. A. Agglomerative Hierarchical Clustering (AHC) of Randomly Amplified Polymorphic DNA (RAPD) Data of Channa punctatus

Gel electrophoresis was performed after the Polymerase Chain Reactions and amplification of the total genomic DNA extracted from the samples of C. punctatus collected from all study sites and the data for scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. To investigate the genetic similarity within and between the populations of C. punctatus collected from the different sites of study in the same cluster/class from the different geographical locations, Jaccard’s coefficient matrix method was used to generate Agglomerative Hierarchical Clustering (AHC) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA) by XLSTAT 2012 version 1.02. A dendrogram showing clustering of genotypes based on amplification of the most scorable bands in random samples and dendrogram of the four separated classes is given in Figure 35 and 36.

The dendrogram developed by this method by the presented data of the scorable bands of the all amplified primers divided the randomly selected individuals of the five populations into four classes/clusters. The division of all the randomly selected five population representative C. punctatus samples collected from different geographical locations in the three cluster was as follows; Blk1, Blk2, Blk3, Blk4, Cha2, Cha3, Cha4, Qbd1, Qbd4, Tsa1, Trm1, Trm3, Trm4, Tsa2, Tsa3 and Tsa4 in first cluster/class, Cha5, Qbd5, Blk5 and Trm2 in second cluster/class, Qbd2, and Trm5 in the third class and Qbd3 and Tsa5 in Fourth class/cluster (Table 103).

The variance decomposition for the optimum classification values remained as 73.93% for within class variation and 26.07% for the between class differences (Table 100). The distance between the class/clusters centroids remained as 0.718 for class 1 and 2, 1.008 for class 1 and 3, 1.008 for class 1 and 4, 1.225 for class 2 and 3 and 2 and 4 while this

270

distance was 1.414 for class 3 and 4 (Table 101). In this classification Cha2 from the Samples collected from Chashma Barrage, Cha5 collected from the same location, Qbd2 collected from Qadirabad Barrage and Qbd3 collected from the same location were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.414 between the central objects of class 1 and 2, 1.000 between the class 1 and 3, 4, respectively, 1.732 between the central objects of class 2 and 3, 4, respectively, while the distance between central objects of class 3 and 4 was 1.414 (Table 102). The results for the conclusion about 4 classes/clusters with their values for within class variance, minimum distance to centroids, average distance to centroids and maximum distance to centroids are given in Table 103.

Figure 35 Dendrogram showing classification of Channa punctatus.

271

272

Figure 36 Dendrogram showing classes of C. punctatus.

273

274

Table 100 Variance decomposition for the optimal classification of C. punctatus

Absolute Percent

Within-class 0.488 73.93% Between-classes 0.172 26.07% Total 0.659 100.00%

Table 101 Distances between the class centroids of C. punctatus

1 2 3 4

1 0 0.718 1.008 1.008

2 0.718 0 1.225 1.225

3 1.008 1.225 0 1.414

275

4 1.008 1.225 1.414 0

Table 102 Central Objects of the Classes of C. punctatus and Distances between them

1 Cha2) 2 Cha5) 3 (Qbd2) 4 (Qbd3)

1 (Cha2) 0 1.414 1.000 1.000

2 (Cha5) 1.414 0 1.732 1.732

3 (Qbd2) 1.000 1.732 0 1.414

4 (Qbd3) 1.000 1.732 1.414 0

Table 103 Results by class of C. punctatus

Class 1 2 3 4

Objects 16 4 2 2

Sum of weights 16 4 2 2

276

Within-class variance 0.250 2.000 0.000 0.000

Minimum distance to centroid 0.125 1.225 0.000 0.000

Average distance to centroid 0.330 1.225 0.000 0.000

Maximum distance to centroid 0.944 1.225 0.000 0.000

Cha2 Cha5 Qbd2 Qbd3

Cha3 Qbd5 Trm5 Tsa5

Cha4 Blk5

Qbd1 Trm2

Qbd4

Blk1

Blk2

Blk3

Blk4

Trm1

Trm3

Trm4

277

Tsa1

Tsa2

Tsa3

Tsa4

B. Agglomerative Hierarchical Clustering (AHC) of Randomly Amplified Polymorphic DNA (RAPD) Data of Channa marulius Gel electrophoresis was performed after the Polymerase Chain Reactions and amplification of the total genomic DNA extracted from the samples of C. marulius collected from all study sites and the data for scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. To investigate the genetic similarity within and between the populations of C. marulius collected from the different sites of study in the same cluster/class from the different geographical locations, Jaccard’s coefficient matrix method was used to generate Agglomerative Hierarchical Clustering (AHC) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA) by XLSTAT 2012 version 1.02. A dendrogram showing clustering of genotypes based on amplification of the most scorable bands in random samples and dendrogram of the four separated classes is given in Figure 37 and 38.

The dendrogram developed by this method by the presented data of the scorable bands of the all amplified primers divided the randomly selected individuals of the five populations into four classes/clusters. The division of all the randomly selected five population representative C. marulius samples collected from different geographical locations in the four clusters was as follows; Blk1, Blk2, Blk3, Blk4, Blk5, Cha1, Cha2, Cha4, Cha5, Qbd1, Qbd2, Qbd3, Qbd5, Trm1, Trm2, Trm3, Trm4, Trm5, Tsa3, Tsa4 and Tsa5 in 1st cluster/class, Cha3 in 2nd cluster/class, Qbd4, and Tsa2 in the 3rd class and Tsa1 in 4th class/cluster (Table 107). 278

The variance decomposition for the optimum classification values remained as 65.90% for within class variation and 34.10% for the between class differences (Table 104). The distance between the class/clusters centroids remained as 1.737 for class 1 and 2, 1.105 for class 1 and 3, 1.009 for class 1 and 4, 1.500 for class 2 and 3, 2.000 for class 2 and 4, while this distance was 1.500 for class 3 and 4 (Table 105). In this classification Cha2 from the Samples collected from Chashma Barrage, Cha3 collected from the same location, Qbd4 collected from Qadirabad Barrage and Tsa1collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.732 between the central objects of class 1 and 2, 1.414 between the class 1 and 3, 1.000 between the central objects of class 1 and 4, 1.732 was the distance between central objects of class 2 and 3, 2.000 was the distance between central objects of class 2 and 4, while this distance for central objects of class 3 and 4 was also 1.732 (Table 106). The results for the conclusion about 4 classes/clusters with their values for within class variance, minimum distance to centroids, average distance to centroids and maximum distance to centroids are given in Table 107.

Figure 37 Dendrogram showing classification of Channa marulius.

279

280

Figure 38 Dendrogram showing classes of C. marulius.

281

282

Table 104 Variance decomposition for the optimal classification of C. marulius

Absolute Percent

Within-class 0.387 65.90%

Between-classes 0.200 34.10%

Total 0.587 100.00%

Table 105 Distances between the class centroids of C. marulius

1 2 3 4

1 0 1.737 1.105 1.009

2 1.737 0 1.500 2.000

3 1.105 1.500 0 1.500

4 1.009 2.000 1.500 0

283

Table 106 Central Objects of the Classes of C. marulius and Distances between them

1 (Cha2) 2 (Cha3) 3 (Qbd4) 4 (Tsa1)

1 (Cha2) 0 1.732 1.414 1.000

2 (Cha3) 1.732 0 1.732 2.000

3 (Qbd4) 1.414 1.732 0 1.732

4 (Tsa1) 1.000 2.000 1.732 0

Table 107 Results by class of C. marulius

Class 1 2 3 4

Objects 21 1 2 1

Sum of weights 21 1 2 1

284

Within-class variance 0.381 0.000 0.500 0.000

Minimum distance to centroid 0.135 0.000 0.500 0.000

Average distance to centroid 0.449 0.000 0.500 0.000

Maximum distance to centroid 0.961 0.000 0.500 0.000

Cha1 Cha3 Qbd4 Tsa1 Cha2 Tsa2 Cha4 Cha5 Qbd1 Qbd2 Qbd3 Qbd5 Blk1 Blk2 Blk3 Blk4 Blk5 Trm1 Trm2 Trm3

285

Trm4 C

Trm5 Ag Tsa3 glo Tsa4 me Tsa5 rati ve Hierarchical Clustering (AHC) of Randomly Amplified Polymorphic DNA (RAPD) Data of Rita rita Gel electrophoresis was performed after the Polymerase Chain Reactions and amplification of the total genomic DNA extracted from the samples of R. rita collected from all study sites and the data for scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. To investigate the genetic similarity within and between the populations of R. rita collected from the different sites of study in the same cluster/class from the different geographical locations, Jaccard’s coefficient matrix method was used to generate Agglomerative Hierarchical Clustering (AHC) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA) by XLSTAT 2012 version 1.02. A dendrogram showing clustering of genotypes based on amplification of the most scorable bands in random samples and dendrogram of the four separated classes is given in Figure 39 and 40.

The dendrogram developed by this method by the presented data of the scorable bands of the all amplified primers divided the randomly selected individuals of the five populations into four classes/clusters. The division of all the randomly selected five population representative R. rita samples collected from different geographical locations in the four clusters was as follows; Blk1, Blk2, Blk3, Blk4, Blk5, Cha1, Cha2, Cha4, Cha5, Qbd1, Qbd2, Qbd3, Qbd5, Trm1, Trm2, Trm3, Trm4, Trm5, Tsa3, Tsa4 and Tsa5 in first cluster/class, Cha3 was the only individual in second cluster/class, Qbd4 and Tsa2 were the two individuals in class 3 while class 4 has only one individual, Tsa1 (Table 111).

286

The variance decomposition for the optimum classification remained as 65.90% for within class variation and 34.10% for the between class differences (Table 108). The distance between the class/clusters centroids remained as 1.737 for class 1 and 2, 1.105 for class 1 and 3, 1.009 for class 1 and 4, 1.500 for class 2 and 3, 2.000 for class 2 and 4, while this distance was 1.500 for class 3 and 4 (Table 109). In this classification Cha2 from the Samples collected from Chashma Barrage, Cha3 collected from the same location, Qbd4 from the samples collected from Qadirabad Barrage and Tsa1 collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.732 between the central objects of class 1 and 2, 1.414 between the central objects of class 1 and 3, 1.000 between the central objects of class 1 and 4, 1.732 was the distance between central objects of class 2 and 3, 2.000 was the distance between central objects of class 2 and 4 while this distance for central objects of class 3 and 4 was 1.732 (Table 110). The results for within class variance, minimum distance to centroids, average distance to centroids and maximum distance to centroids are given in Table 111.

Figure 39 Dendrogram showing classification of R. rita.

287

288

Figure 40 Dendrogram showing classes of R. rita.

289

290

Table 108 Variance decomposition for the optimal classification of Rita rita

Absolute Percent

Within-class 0.387 65.90%

Between-classes 0.200 34.10%

Total 0.587 100.00%

Table 109 Distances between the class centroids of R. rita

1 2 3 4

1 0 1.737 1.105 1.009

2 1.737 0 1.500 2.000

291

3 1.105 1.500 0 1.500

4 1.009 2.000 1.500 0

Table 110 Central Objects of the Classes of R. rita and Distances between them

1 (Cha2) 2 (Cha3) 3 (Qbd4) 4 (Tsa1)

1 (Cha2) 0 1.732 1.414 1.000

2 (Cha3) 1.732 0 1.732 2.000

3 (Qbd4) 1.414 1.732 0 1.732

4 (Tsa1) 1.000 2.000 1.732 0

292

Table 111 Results by class of Rita rita

Class 1 2 3 4

Objects 21 1 2 1

Sum of weights 21 1 2 1

Within-class variance 0.381 0.000 0.500 0.000

Minimum distance to centroid 0.135 0.000 0.500 0.000

Average distance to centroid 0.449 0.000 0.500 0.000

Maximum distance to centroid 0.961 0.000 0.500 0.000 Cha1 Cha3 Qbd4 Tsa1 Cha2 Tsa2 Cha4 Cha5 Qbd1 Qbd2 Qbd3 Qbd5 Blk1 Blk2 Blk3

293

Blk4 Blk5 Trm1 Trm2 Trm3 Trm4 Trm5 Tsa3 Tsa4 Tsa5

D Agglomerative Hierarchical Clustering (AHC) of Randomly Amplified Polymorphic DNA (RAPD) Data of Sperata seenghala Gel electrophoresis was performed after the Polymerase Chain Reactions and amplification of the total genomic DNA extracted from the samples of S. seenghala collected from all study sites and the data for scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. To investigate the genetic similarity within and between the populations of S. seenghala collected from the different sites of study in the same cluster/class from the different geographical locations, Jaccard’s coefficient matrix method was used to generate Agglomerative Hierarchical Clustering (AHC) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA) by XLSTAT 2012 version 1.02. A dendrogram showing clustering of genotypes based on amplification of the most scorable bands in random samples and dendrogram of the four separated classes is given in Figure 41 and 42.

294

The dendrogram developed by the presented data of the scorable bands of all amplified primers divided the randomly selected individuals of the five populations into four classes/clusters in case of S. seenghala. The division was as follows; Blk1, Blk2, Blk3, Blk4, Blk5, Cha1,Cha2, Cha3, Cha4, Cha5, Qbd2, Qbd3, Qbd4, Trm2, Trm3, Trm4, Trm5, Tsa3, Tsa4 and Tsa5 in first cluster/class, Qbd1 and Tsa2 in second cluster/class, Class/cluster 3 and 4 contains only one individual each, Trm1 and Tsa1, respectively (Table 115).

The variance decomposition for the optimum classification values remained as 64.19% within class variation and 35.81% between class differences (Table 112). The distance between the class/clusters centroids remained as 1.233 for class 1 and 2, 1.421 for class 1 and 3, 1.682 for class 1 and 4, 1.871 for class 2 and 3, 2.121 for class 2 and 4, while this distance was 2.236 for class 3 and 4 (Table 113). In this classification Cha2 from the Samples collected from Chashma Barrage, Qbd1 from the samples collected from Qadirabad Barrage and Trm1 collected from Trimu barrage at the junction of Chenab and Jhelum rivers and Tsa1 collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.414 between the central objects of class 1 and 2, 3 respectively, 1.732 between the central objects of class 1 and 4, 2.000 was the distance between central objects of class 2 and 3, while this distance for central objects of class 2 and 4 and 3 and 4 was 2.236, respectively (Table 114). The results for the conclusion about 4 classes/clusters with their values for within class variance, minimum distance to centroids, average distance to centroids and maximum distance to centroids are given in Table 115.

Figure 41 Dendrogram showing classification of S. seenghala.

295

296

Figure 42 Dendrogram showing classes of S. seenghala.

297

298

Table 112 Variance decomposition for the optimal classification of S. seenghala

Table 113 Distances between the class centroids of S. seenghala Table 114 Central Objects of the Classes of S. seenghala and DistancesAbsolute Percent Withinbetween-class them 1 2 0.456 3 64.19%4 Between1 -classes 0 1.233 0.254 1.421 35.81%1.682 1 (Cha2) 2 (Qbd1) 3 (Trm1) 4 (Tsa1) Total 2 1.233 0 0.710 1.871 100.00%2.121 31 (Cha2) 1.421 0 1.8711.414 0 1.414 2.2361.732 4 1.682 2.121 2.236 0 2 (Qbd1) 1.414 0 2.000 2.236

3 (Trm1) 1.414 2.000 0 2.236

4 (Tsa1) 1.732 2.236 2.236 0

Table 115 Results by class of S. seenghala

299

Class 1 2 3 4

Objects 21 2 1 1

Sum of weights 21 2 1 1

Within-class variance 0.429 1.000 0.000 0.000

Minimum distance to centroid 0.143 0.707 0.000 0.000

Average distance to centroid 0.473 0.707 0.000 0.000

Maximum distance to centroid 1.353 0.707 0.000 0.000

Cha1 Qbd1 Trm1 Tsa1 Cha2 Tsa2 Cha3 Cha4 Cha5 Qbd2 Qbd3 Qbd4 Qbd5 Blk1 Blk2 Blk3

300

Blk4 Blk5 Trm2 Trm3 Trm4 Trm5 Tsa3 Tsa4 Tsa5

E Agglomerative Hierarchical Clustering (AHC) of Randomly Amplified Polymorphic DNA (RAPD) Data of Wallago attu Gel electrophoresis was performed after the Polymerase Chain Reactions and amplification of the total genomic DNA extracted from the samples of W. attu collected from all study sites and the data for scorable amplified bands was organized on the one-zero pattern i.e. the presence of band recorded as “1” and for absence marked as “0”. To investigate the genetic similarity within and between the populations of W. attu collected from the different sites of study in the same cluster/class from the different geographical locations, Jaccard’s coefficient matrix method was used to generate Agglomerative Hierarchical Clustering (AHC) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA) by XLSTAT 2012 version 1.02. A dendrogram showing clustering of genotypes based on amplification of the most scorable bands in random samples and dendrogram of the three separated classes is given in Figure 43 and 44.

301

The dendrogram developed by this method by the presented data of the scorable bands of the all amplified primers divided the randomly selected individuals of the five populations into six classes/clusters. The division of all the randomly selected five population representative W. attu samples collected from different geographical locations in the four clusters was as follows; Blk1, Blk3, Cha1, Cha4, Cha5, Qbd2, Qbd3, Qbd4, Trm1, Trm2, Trm3, Trm4, Trm5, Tsa2, Tsa4 and Tsa5 in first cluster/class, Blk2, Cha2 and Tsa3 was the only individual in second cluster/class, Cha3 was the only individual in class 3, Class/cluster 4 consists of Blk5, Qbd1 and Tsa1 while Class 5 and 6 was consists of one individual each, Qbd5 and Blk4, respectively (Table 119).

The variance decomposition for the optimum classification values remained as 49.14% for within class variation and 50.86% for the between class differences (Table 116). The distance between the class/clusters centroids remained as 1.116 for class 1 and 2, 1.422 for class 1 and 3, 1.065 for class 1 and 4, 1.378 for class 1 and 5, 1.422 for class 1 and 6, 1.795 for class 2 and 3, 1.528 for class 2 and 4, 1.795 for class 2 and 5, 6; 1.764 for class 3 and 4, 2.000 for class 3 and 5,6; 1.764 for class 4 and 5,6 while this distance was 2.000 for class 5 and 6 (Table 117). In this classification Cha5 from the Samples collected from Chashma Barrage, Blk2 collected from the Baloki Barrage At River Ravi, Cha3 from the Samples collected from Chashma Barrage, Qbd1 and Qbd5 from the samples collected from Qadirabad Barrage, while Blk4 collected from the Baloki Barrage at River Ravi were central objects for class/clusters 1, 2, 3, 4, 5, and 6, respectively. The central objects of the classes and the distance between them are shown in Table 118. The results for the conclusion about 4 classes/clusters with their values for within class variance, minimum distance to centroids, average distance to centroids and maximum distance to centroids are given in Table 119.

Figure 43 Dendrogram showing classification of W. attu.

302

303

Figure 44 Dendrogram showing classes of W. attu.

304

305

Table 116 Variance decomposition for the optimal classification of W. attu Absolute Percent Within-class 0.401 49.14% Between-classes 0.415 50.86% Total 0.817 100.00%

Table 117 Distances between the class centroids of W. attu

1 2 3 4 5 6

1 0 1.116 1.422 1.065 1.378 1.422

2 1.116 0 1.795 1.528 1.795 1.795

3 1.422 1.795 0 1.764 2.000 2.000

4 1.065 1.528 1.764 0 1.764 1.764

5 1.378 1.795 2.000 1.764 0 2.000

306

6 1.422 1.795 2.000 1.764 2.000 0

Table 118 Central Objects of the Classes of W. attu and Distances

between them

Table 119 Results1 (Cha5)by class 2of (Bl W.k2) attu 3 (Cha3) 4 (Qbd1) 5 (Qbd5) 6 (Blk4)

1 (Cha5) 0 1.000 1.414 1.000 1.414 1.414 Class 1 2 3 4 5 6 2 (Blk2) 1.000 0 1.732 1.414 1.732 1.732 Objects 16 3 1 3 1 1 3 (Cha3) 1.414 1.732 0 1.732 2.000 2.000 Sum of weights 16 3 1 3 1 1 4 (Qbd1) 1.000 1.414 1.732 0 1.732 1.732 Within-class variance 0.375 0.667 0.000 0.333 0.000 0.000 5 (Qbd5) 1.414 1.732 2.000 1.732 0 2.000 Minimum distance to centroid 0.153 0.471 0.000 0.333 0.000 0.000 6 (Blk4) 1.414 1.732 2.000 1.732 2.000 0 Average distance to centroid 0.451 0.654 0.000 0.444 0.000 0.000

Maximum distance to centroid 0.948 0.745 0.000 0.667 0.000 0.000

Cha1 Cha2 Cha3 Qbd1 Qbd5 Blk4

Cha4 Blk2 Blk5

Cha5 Tsa3 Tsa1

Qbd2

307

Qbd3 4.7 Principle Component Qbd4 Analysis (PCA) of Randomly Amplified Polymorphic DNA Blk1 (RAPD) Data of Experimental Blk3 Species

Trm1 The data obtained from the random

Trm2 markers for the Randomly Amplified Polymorphic DNA (RAPD) was Trm3 subjected to the XLSTAT 2012 Trm4 version 1.02, computer software. Trm5 The Principal Component Analysis (PCA) for grouping of the different Tsa2 genotypes from the different Tsa4 environmental conditions was done Tsa5 by Spearman Varimax rotation method for bi-plot generation of the co-occurrence of the same genotypes with similar genetic properties and specificity of different primers in the same program.

A Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) Data of Channa punctatus

308

The Spearman Varimax rotation method of Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 to make conclusive results about the genetic relationships among the collected samples of C. punctatus from different geographical locations along with differentiation between and within the groups.

The results obtained from the PCA indicated clearly that the increase in the number of factors or components is correlated with the decrease in Eigen values (Table-120). In the table values are at its maximum in the level of first and second factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first four main factors accounted for 55.849% of cumulative variability. Therefore, we can assume after observing the results that the PCA grouped the tested variables or parameters of the fish RAPD amplification data into four main components which all together accounted for 55.85% of the cumulative variation among the factors. The first and second group (F1 and F2) amongst the major four groups accounted for 14.881% each, of the variability percentage while the third and fourth (F3 and F4) from these accounted for 14.069% and 12.018% respectively, of the cumulative variability. The values are 7.764%, 7.605% for fifth and sixth factor, respectively, 7.440% for seventh, eighth and ninth, each. These values are 4.310% and 2.151% for F10 and F11 respectively. The scree plot represents the trends of the factors in Figure-45.

The Figure-46 showed the trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into four major variable groups, one group towards the positive side and one group towards the negative side and two groups had the neutral affect. In the Figure 47 it is clearly indicated that the representative individuals of five populations on the basis of two major factors are accumulated as a one group. Figure-48 showed the bi- plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations.

Table 120 Eigen values of RAPD data (PCA) for C. punctatus

309

F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Eigenvalue 2.083 2.083 1.970 1.683 1.087 1.065 1.042 1.042 1.042 0.603 0.301 Variability (%) 14.881 14.881 14.069 12.018 7.764 7.605 7.440 7.440 7.440 4.310 2.151 Cumulative % 14.881 29.762 43.831 55.849 63.613 71.218 78.659 86.099 93.540 97.849 100.000

Figure 45 Graph between Eigen values and cumulative variability for Channa punctatus

Figure 46 Graph between variables and Factors for C. punctatus

310

Figure 47 Graph between Observations and factors for C. punctatus

311

Figure 48 Biplot graph between F factors for C. punctatus

312

Results after varimax rotation for Channa punctatus

After the varimax rotation which is the alteration in axes in the PCA which increases the cumulative variability of the squared loading. This is the orthogonal rotation which is used to show the influence or share of each individual. According to the Kaiser (1958) criterion, this may be the rotation which clears the individuals on such a level that "for each factor, high loadings (correlations) will result for a few variables; the rest will be near zero." The varimax rotation criterion maximizes the sum of the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. The table-121 shows the Rotation matrix for C. punctatus and Table 122 showed the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here,

313

now the cumulative variance is same as before rotation as 55.849% with some variation in individual factor variation which remained as 14.881% and 14.881% for D1 and D2, each.

The Figure-49 showed the trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into four major variable groups, one group towards the positive side, one towards negative and two groups at the neutral site. The Figure-50 indicated the individuals of five populations taken as observations and divided into two major factors after varimax rotation. In this picture it is clearly indicated that the representative individuals of five populations are genetically correlated with each other with some exceptional case like Cha5 sample from Chashma Barrage towards somewhat negative value and Blk5 sample from the Baloki Barrage of the River Ravi towards the extreme negative distinction. The Trm2 is at extreme positive distinction. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. Figure-51 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations which have been minimizes after varimax rotation.

Table 121 Rotation matrix for C. punctatus

D1 D2

D1 -1.000 0.000

D2 0.000 -1.000

Table 122 Percentage of variance after Varimax rotation for C. punctatus

314

D1 D2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Variability (%) 14.881 14.881 14.069 12.018 7.764 7.605 7.440 7.440 7.440 4.310 2.151

Cumulative % 14.881 29.762 43.831 55.849 63.613 71.218 78.659 86.099 93.540 97.849 100.000

Figure 49 Graph between variables after Varimax rotation for C. punctatus

315

Figure 50 Graph between observations after Varimax rotation for C. punctatus

316

Figure 51 Biplot graph between F factors after Varimax rotation for C. punctatus

317

B Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) Data of Channa marulius.

The Spearman Varimax rotation method of Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 to make conclusive results about the genetic relationships among the collected samples of C. marulius from different geographical locations along with differentiation between and within the groups.

The results obtained from the PCA indicated clearly that the increase in the number of factors or components was correlated with the decrease in Eigen values (Table-123). The values in the table showed that the values were at its maximum at the level of first factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first three main factors accounted for 42.770% of cumulative variability. Therefore, we can assume after 318

observing the results that the PCA grouped the tested variables or parameters of the fish RAPD amplification data into three main components which all together accounted for 42.78% of the cumulative variation among the factors. The first, second and third group (F1, F2 and F3) amongst the major three groups accounted for 16.162%, 14.415%, 12.193%, each, of the cumulative variability while the fourth to ninth (F4 to F9) from these accounted for 8.013% each and tenth and eleventh (F10 and F11) 5.091 and 4.062% respectively, of the cumulative variability. The scree plot represents the trends of the factors in Figure-52.

The Figure-53 showed the trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into five major variable groups’ two groups towards the positive side and two groups towards the negative side with one at neutral place. In the Figure 54 it is clearly indicated that the representative individuals of five populations on the basis of two major factors are accumulated as a one group. Figure-55 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations.

Table 123 Eigen values of RAPD data (PCA) for C. marulius

F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Eigenvalue 2.101 1.874 1.585 1.042 1.042 1.042 1.042 1.042 1.042 0.662 0.528

Variability (%) 16.162 14.415 12.193 8.013 8.013 8.013 8.013 8.013 8.013 5.091 4.062

Cumulative % 16.162 30.577 42.770 50.783 58.796 66.809 74.821 82.834 90.847 95.938 100.000

319

Figure 52 Graph between Eigen values and cumulative variability for C. marulius

320

Figure 53 Graph between variables and Factors for C. marulius

321

Figure 54 Graph between Observations and factors for C. marulius

322

Figure 55 Biplot graph between F factors for C. marulius

323

Results after varimax rotation for Channa marulius

The varimax rotation which is the alteration in axes in the PCA increases the cumulative variability of the squared loading. This is the orthogonal rotation which is used to show the influence or share of each individual. According to the Kaiser (1958) criterion, this may be the rotation which clears the individuals on such a level that "for each factor, high loadings (correlations) will result for a few variables; the rest will be near zero." The varimax rotation criterion maximizes the sum of

324

the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. The Table-124 shows the Rotation matrix after Varimax rotation for C. marulius shows The Table-125 shows the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here, now the cumulative variance is same as before rotation as 30.577% with some variation in individual factor variation which remained as 15.580% and 14.997% for D1 and D2, respectively.

The Figure-56 showed the trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into five major variable groups, three groups towards the negative side and one group towards the positive side while the other group has neutral effect. The Figure-57 indicated the individuals of five populations taken as observations and divided into two major factors after varimax rotation. In this picture it is clearly indicated that the representative individuals of five populations are genetically positively correlated with each other with some exceptional cases like Cha3 sample from the Chashma Barrage of the River Indus towards the extreme negative value. And Tsa1, Qbd2 and Trm1 are at extreme positive. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. Figure-58 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations which have been minimizes after varimax rotation.

Table 124 Rotation matrix after Varimax rotation for C. marulius

D1 D2 D1 0.817 -0.577

325

D2 0.577 0.817

Table 125Percentage of variance after Varimax rotation C. marulius

D1 D2 F3 F4 F5 F6 F7 F8 F9 F10 F11 Variability (%) 15.580 14.997 12.193 8.013 8.013 8.013 8.013 8.013 8.013 5.091 4.062

Cumulative % 15.580 30.577 42.770 50.783 58.796 66.809 74.821 82.834 90.847 95.938 100.000

Figure 56 Graph between variables after Varimax rotation for C. marulius

326

Figure 57 Graph between observations after Varimax rotation for C. marulius

327

Figure 58 Biplot graph between F factors after Varimax rotation for C. marulius

328

Values in bold correspond for each observation to the factor for which the squared cosine is the largest

329

C Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) Data of for Rita rita

The Spearman Varimax rotation method of Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 to make conclusive results about the genetic relationships among the collected samples of R. rita from different geographical locations along with differentiation between and within the groups.

The results obtained from the PCA indicated clearly that the increase in the number of factors or components is correlated with the decrease in Eigen values (Table- 126). In the table values are at its maximum in the level of first and second factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first two main factors accounted for 30.968% of cumulative variability. Therefore, we can assume after observing the results that the PCA grouped the tested variables or parameters of the fish RAPD amplification data into two main components which all together accounted for 30.97% of the cumulative variation among the factors. The first and second group (F1 and F2) amongst the major two groups accounted for 20.609% and 10.359%, respectively, of the cumulative variability while the third to ninth (F3 to F9) from these accounted for 8.681% each respectively, of the cumulative variability. The scree plot represents the trends of the factors in Figure 59.

The Figure-60 showed the trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into five major variable groups three groups towards the positive side and one group towards the negative side while the remaining at the neutral distinction. In the Figure 61, it is clearly indicated that the representative individuals of five populations on the basis of two major factors are accumulated as a one group. Figure 62 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations. 330

Table 126 Eigen values of RAPD data (PCA) for Rita rita

F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Eigenvalue 2.473 1.243 1.042 1.042 1.042 1.042 1.042 1.042 1.042 0.613 0.380

Variability (%) 20.609 10.359 8.681 8.681 8.681 8.681 8.681 8.681 8.681 5.106 3.163

Cumulative % 20.609 30.968 39.649 48.329 57.010 65.690 74.371 83.051 91.732 96.837 100.000

Figure 59 Graph between Eigen values and cumulative variability for R. rita

331

Figure 60 Graph between variables and Factors for R. rita

332

Figure 61 Graph between Observations and factors for R. rita

333

Figure 62 Biplot graph between F factors for R. rita

334

Results after varimax rotation for Rita rita

According to the Kaiser (1958) criterion, this may be the rotation which clears the individuals on such a level that "for each factor, high loadings (correlations) will result for a few variables; the rest will be near zero." The varimax rotation criterion maximizes the sum of the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. The table-127 shows the Rotation matrix for C. punctatus. The table 126 showed the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here, now the cumulative variance is same as before rotation as 30.968% with some variation in individual factor variation which remained as 20.141% and 10.827% for D1 and D2, respectively.

335

The Figure 63 showed the trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into five major variable groups, two groups towards the positive side and one group towards the negative side and remaining at the neutral place. The Figure 64 indicated the individuals of five populations taken as observations and divided into two major factors after varimax rotation. In this picture it is clearly indicated that the representative individuals of five populations are genetically positively correlated with each other with some exceptional case like Tsa2 sample from Taunsa Barrage towards somewhat negative value and Cha3 sample from the Chashma Barrage of the River Indus towards the extreme negative value. The sample Qbd4 and Cha1 are at extreme positive limit. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. Figure-65 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations which have been minimizes after varimax rotation.

Table 127 Rotation matrix for R. rita

D1 D2

D1 0.977 0.214

D2 -0.214 0.977

Table 128 Percentage of variance after Varimax rotation for R. rita

336

D1 D2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Variability 20.141 10.827 8.681 8.681 8.681 8.681 8.681 8.681 8.681 5.106 3.163 (%)

Cumulative 20.141 30.968 39.649 48.329 57.010 65.690 74.371 83.051 91.732 96.837 100.000 %

Figure 63 Graph between variables after Varimax rotation for R. rita

Figure 64 Graph between observations after Varimax rotation for R. rita

337

Figure 65 Biplot graph between F factors after Varimax rotation for R. rita

338

D Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) Data of for S. seenghala

The Spearman Varimax rotation method of Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 to make conclusive results about the genetic relationships among the for S. seenghala collected samples from different geographical locations along with differentiation between and within the groups.

The results obtained from the PCA indicated clearly that the increase in the number of factors or components was correlated with the decrease in Eigen values (Table-129). The values in the table showed that the values were at their maximum at the level of first and second factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first four main factors accounted for 55.406% of cumulative variability. Therefore, we can assume after observing the results that the PCA grouped the tested variables or parameters of the fish RAPD amplification data into four main components which all together accounted for 55.41% of the cumulative variation among

339

the factors. The first and second group (F1 and F2) amongst the major four groups accounted for 15.717% and 14.044% respectively, of the cumulative variability while the third and fourth (F3 and F4) from these accounted for 13.889% and 11.756% respectively, of the cumulative variability. The factors fifth to ninth accounted for 6,944% each. The factors tenth, eleventh and twelfth (F10, F11 and F12) accounted for 4.815%, 3.623% and 1.433% respectively. The scree plot represents the trends of the factors in Figure-66.

The Figure-67 showed the trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into seven major variable groups’ two groups towards the positive side and four groups towards the negative side and one has neutral distinction. The Figure-68 clearly indicated that the representative individuals of five populations on the basis of two major factors are accumulated as a one group. Figure-69 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations.

Table 129 Eigen values of RAPD data (PCA) for S. seenghala

F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12

Eigenvalue 2.358 2.107 2.083 1.763 1.042 1.042 1.042 1.042 1.042 0.722 0.543 0.215

Variability (%) 15.717 14.044 13.889 11.756 6.944 6.944 6.944 6.944 6.944 4.815 3.623 1.433

Cumulative % 15.717 29.761 43.650 55.406 62.351 69.295 76.240 83.184 90.129 94.944 98.567 100.000

Figure 66 Graph between Eigen values and cumulative variability for S. seenghala

340

Figure 67 Graph between variables and Factors for S. seenghala

341

Figure 68 Graph between Observations and factors for S. seenghala

342

Figure 69 Biplot graph between F factors for S. seenghala

343

Results after varimax rotation for Sperata seenghala

In the PCA after varimax rotation which is the alteration in axes increases the cumulative variability of the squared loading. This is the orthogonal rotation which is used to show the influence or share of each individual. According to the Kaiser (1958) criterion, this may be the rotation which clears the individuals on such a level that "for each factor, high loadings (correlations) will result for a few variables; the rest will be near zero." The varimax rotation criterion maximizes the sum of the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. The table-130 shows the Rotation matrix for S. seenghala after varimax rotation. The table-131 shows the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here, now the cumulative variance is same as before rotation as 29.761% with some variation in individual factor variation which remained as 15.694% and 14.067% for D1 and D2, respectively.

344

The Figure-70 showed the trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into seven major variable groups, two groups towards the positive side and four groups towards the negative side and one at the neutral affect. The Figure-71 indicated the individuals of five populations taken as observations and divided into two major factors after varimax rotation. In this picture it is clearly indicated that the representative individuals of five populations are genetically positively correlated with each other with some exceptional case like Qbd3 and Blk3 with somewhat negative value and Tsa1 with extreme negative value along with Qbd1 and Tsa2 and trm1 and Cha3 at extreme positive places. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. Figure-72 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations which have been minimizes

Table 130 Rotation matrix for S. seenghala

D1 D2

D1 0.993 0.116

D2 0.116 -0.993

Table 131 Percentage of variance after Varimax rotation for S. seenghala

D1 D2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12

Variability (%) 15.694 14.067 13.889 11.756 6.944 6.944 6.944 6.944 6.944 4.815 3.623 1.433

345

Cumulative % 15.694 29.761 43.650 55.406 62.351 69.295 76.240 83.184 90.129 94.944 98.567 100.000

Figure 70 Graph between variables after Varimax rotation for S. seenghala

346

Figure 71 Graph between observations after Varimax rotation for S. seenghala

347

Figure 72 Biplot graph between F factors after Varimax rotation for S. seenghala

348

E Principle Component Analysis (PCA) of Randomly Amplified Polymorphic DNA (RAPD) Data of Wallago attu

The Spearman Varimax rotation method of Principal Component Analysis (PCA) was conducted by XLSTAT 2012 version 1.02 to make conclusive results about the genetic relationships among the W. attu collected samples from different geographical locations along with differentiation between and within the groups.

The results obtained from the PCA indicated clearly that the increase in the number of factors or components was correlated with the decrease in Eigen values (Table-132). The values in the table were at its maximum at the level of first and second factor. In the same way according to the Kaiser (1958) criterion based upon the Eigen values greater than one, first four main factors accounted for 50.369% of cumulative variability. Therefore, we can assume after observing the results that the PCA grouped the tested variables or parameters of the fish RAPD amplification data into four main

349

components which all together accounted for 50.37% of the cumulative variation among the factors. The first and second group (F1 and F2) amongst the major four groups accounted for 13,043% each, of the cumulative variability while the third and fourth (F3 and F4) from these accounted for 12.764% and 11.544% respectively, of the cumulative variability. The fifth factor (F5) accounted for 9.154% while the factors sixth to tenth (F6 to F10) from these accounted for 6.522% each. Factor eleventh, twelfth and thirteenth (F11, F12 and F13) 4.043%, 2.588% and 1.211%, respectively, of the cumulative variability. The scree plot represents the trends of the factors in Figure-73.

The Figure-74 showed the trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into four major variable groups, one group towards the positive side, one groups toward the negative side and two groups has the neutral distinction. The Figure 75 clearly indicates that the representative individuals of five populations on the basis of two major factors are accumulated as a one group. Figure-76 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations.

Table 132 Eigen values of RAPD data (PCA) for W. attu

F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 F13

Eigenvalue 2.083 2.083 2.037 1.842 1.468 1.042 1.042 1.042 1.042 1.042 0.665 0.419 0.194

Variability 13.021 13.021 12.734 11.513 9.174 6.510 6.510 6.510 6.510 6.510 4.156 2.618 1.211 (%)

Cumulative 13.021 26.042 38.776 50.289 59.463 65.973 72.484 78.994 85.505 92.015 96.171 98.789 100.000 %

Figure 73 Graph between Eigen values and cumulative variability for W. attu 350

Figure 74 Graph between variables and Factors for W. attu

351

Figure 75 Graph between Observations and factors for W. attu

352

Figure 76 Biplot graph between F factors for W. attu

353

Results after varimax for Wallago attu

After the varimax rotation which is the alteration in axes in the PCA the cumulative variability of the squared loading is increased. This is the orthogonal rotation which is used to show the influence or share of each individual. According to the Kaiser (1958) criterion, this may be the rotation which clears the individuals on such a level that "for each factor, high loadings (correlations) will result for a few variables; the rest will be near zero." The varimax rotation criterion maximizes the sum of the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. The table-133 shows the rotation matrix after Varimax rotation for W. attu The table-134 shows the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here,

354

now the cumulative variance is same as before rotation as 26.087% with some variation in individual factor variation which remained as 13.043%, each.

The Figure-77 showed the trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into four major variable groups, one group towards the positive side and one group towards the negative side and the two with neutral affect. The Figure-78 indicated the individuals of five populations taken as observations and divided into four major factors after varimax rotation. In this picture it is clearly indicated that the representative individuals of five populations are genetically correlated with each other with some exceptional case like Qbd5 with some positive correlation and Cha3 with extreme positive value and Blk4 sample from the Baloki Barrage of the River Ravi towards the extreme negative value. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. Figure-79 showed the bi-plot analysis of the variables (primers) and observations (representative individuals of the five populations), which indicates the level of similarity and differences among the five populations which have been minimizes after varimax rotation.

Table 133 Rotation matrix for W. attu

D1 D2

D1 1.000 0.000

D2 0.000 -1.000

Table 134 Percentage of variance after Varimax rotation for W. attu

355

D1 D2 F3 F4 F5 F6 F7 F8 F9 F10 F11 F12 F13 F14

Variability (%) 13.043 13.043 12.764 11.544 9.154 6.522 6.522 6.522 6.522 6.522 4.043 2.588 1.211 13.043

Cumulative % 13.043 26.087 38.851 50.396 59.549 66.071 72.593 79.114 85.636 92.158 96.201 98.789 100.000 13.043

Figure 77 Graph between variables after Varimax rotation for W. attu

356

Figure 78 Graph between observations after Varimax rotation for W. attu

357

Figure 79 Biplot graph between F factors after Varimax rotation for W. attu

358

359

Chapter 5

DISCUSSION

The study on Genetic variability in some commercially important carnivorous fishes by molecular markers in Punjab, Pakistan was conducted for comparison of the natural genetic populations of carnivorous fishes for intra specific variability for investigation of the phylogenetic relationship and the study of level of heterogeneticity amongst the natural populations on species level. Morphometric Parameters In the present study of morphometric parameters, the data subjected to Minitab computer software for the ANOVA. The data regarding the wet body weight and total length of C. punctatus and S. seenghala were significantly different whereas the weight of C. marulius, R. rita and W. attu were non significantly different among the different geographical locations when analysis was made. The results of variance of fork length of S. seenghala were significant, while the results of variance for fork length of C. marulius, C. punctatus, R. rita and W. attu were non significantly different among the different geographical locations. The results of morphometric parameter for head length by the analysis of variance showed that it was significantly different for C. punctatus and for S. seenghala. Head length of C. marulius, R. rita and W. attu were none significantly different among the different geographical locations. In the analysis for the stoutness, another variance of the morphometric parameter, for the S. Seenghala the difference was significant, while the stoutness of C. marulius, C. punctatus, R. rita and W. attu were non-significantly different among the different geographical locations. The outcomes by the analysis of variance of the morphometric parameter from the data obtained for the dorsal fin length for all the fishes, showed that the dorsal fin length of S. seenghala were significant, while the dorsal fin length of C. marulius, C. punctatus, R. rita and W. attu were non significantly different among the different geographical locations. The data

360

obtained for the length of the caudal fin for all the fishes when analyzed, the variance showed that there was significant difference (p < 0.05) among the samples of the C. punctatus and S. seenghala for different geographical sites. The figures obtained for the C. marulius, R. rita and W. attu were non-significantly different. The facts obtained for the length of the anal fin showed that the anal fin length of S. seenghala were significant, while the anal fin length of other experimental fishes was non-significantly different. The findings by the analysis of variance for the adipose fin for R. rita were significantly different among the different sampling locations in the study. The data collected for the length of the Pectoral fin for all the fishes got analyzed and the result of variance showed that there was significant difference (p < 0.05) among the C. punctatus and S. seenghala for different geographical sites. The data obtained for the C. marulius, R. rita and W. attu was non-significantly different. The results by the analysis of variance of the morphometric parameters showed that the average pectoral fins length of C. punctatus was significant, weight of C. marulius, R. rita and W. attu were non significantly different while the average pectoral fins length of S. seenghala was significantly different among the different geographical locations. These results are in accordance with the studies of DARS et al., (2012) while working with morphometric and meristic characters and their relationship in C. punctatus collected from River Indus near District Jamshoro of the Sindh Province in Pakistan. Their results of the study were consisted on the observation for male and female individuals of the target species separately. They concluded that the standard length, lateral line length, tail fin length, dorsal fin length, pectoral fin length, girth and head length of the C. punctatus were linearly correlated with the total body length, while ventral fin, pelvic fin length and gap of mouth parameters showed high correlation with the total length of the body and head length especially for the male individuals of the said species. While their results for the female individuals of the target species showed that standard length, lateral line length, pelvic fin length, dorsal fin length, pectoral fin length and head length found to be linearly correlated. But the parameters including girth and tail fin length were found to be highly correlated with the total length of female fish of the target species. Same results were found in the studies of SAIKIA

361

(2012) in C. punctatus while working in Sivsagar, Assam, India. These results are also in line with the findings of PERVAIZ et al., (2012) while working with the T. macrolepis collected from different locations of Attock District and adjoining areas. They observed that high level of significant relationships were observed with total length (TL) and head length (HL) when compared to all other morphometric parameters. The results coherent with the present study were also concluded by NAREJO et al., (2008) while working with the two different types of the Tenualosa ilisha collected from River Indus for their morphometric and meristic differences. Such studies were also conducted on fluvial Japanese Char by Nakamura (2003) for their meristic and morphometric variations. For the morphometric and genetic variation, studies were also conducted by POULET et al., (2004) on Sander lucioperca (Pike perch) on the same line the work was conducted by LASHARI et al., (2004) with the morphometric characters and their relationship in Cirrhinus reba. Same type of studies for morphometric and genetic structures of Liza abu populations collected from Rivers Orontes, Euphrates and Tigris were performed by TURAN et al., (2004). Morphometric comparison in Giant river catfish (Mystus seenghala) has been made by SAINI et al. (2008) for two places from rivers of Indus river system from the Beas river compared with a population in the Sutlej River of the Indus river system using 28 morphometric characters. Allometric transformation of each measurement was done to eliminate correlations with size. They performed the stepwise discriminant analysis and thus retained nine variables that significantly discriminated the Beas samples from the Sutlej samples. Using these variables, 91.2% (original) and 89.0% (cross validated) of fish is classified into their correct samples. Misclassification was higher for the Sutlej samples (12.5%) than for the Beas samples (6.3%). The results of their study for discriminant analyses showed that variability in the Beas samples was more homogeneous and provided a more characteristic picture of the group than the Sutlej samples. The univariate ANOVA revealed significant differences between the means of the two populations for 12 of the 28 transformed morphometric measurements. Also in our study some morphometric characteristic has shown the same patterns.

362

The results of ANOVA for most of the parameters in Channa punctatus, C. marulius, Rita rita and W. attu are non- significantly different in present study and these results are comparable to the findings of GARG et al. (2010).The random amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR) was applied to analyze the genetic variation of the 2 populations of Mystus vittatus (Bloch) of Madhya Pradesh, India. Individuals collected from 2 sites, namely Bhadwada reservoir (Bhopal) and Mohinisagar reservoir (Gwalior). These primers produced 388 scorable DNA fragments were found, of which 252 (64.98%) were polymorphic, 182 (46.90%) were monomorphic and 14 (3.61%) were unique. RAPD banding patterns, showed variations between and within the populations, while, the morphological variations were negligible.

In present study, morphometric data for Channa punctatus was subjected to Principal Component Analysis (PCA) and the cumulative variability for first two main factors accounted for 98.706% and among them the first group accounted for 84.415% of the variability while the second accounted for 14.291%. The variability percentage for factor F3 and F4, was for less as the values were 0.825% and 0.469%, respectively The correlation of F1 and F2 for all the parameters was positive except for dorsal fin for F2. A positive correlation for wet body weight, total length, dorsal fin length and for wet body weight, total length, dorsal fin length average length of paired pectoral fin was observed for F3 and F4, respectively The observation plot and bi-plot of the morphometric parameters showed that the genotypes were divided into four classes by the cluster analysis were also in the same groups. In the Bartlett's sphericity test for C. punctatus, the critical Chi-square value was 32.671. Squared cosine values were largest for different parameters in factor F1 except the wet body weight of the fish for which the value was larger for the factor F2.

According to the Kaiser (1958) Criterion based upon the Eigen values greater than one, The cumulative variability was 99.996% for the first two main factors Therefore, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of C. marulius morphometery into two main components, The first group amongst the

363

major two groups accounted for 99.849% of the cumulative variability while the second accounted for 0.148% of the cumulative variability. Factor three (F3) has a value of 0.003% only. The factor one (F1) had a positive correlation for all the studied parameters whereas the factor two (F2) were negatively correlated with all the parameters except dorsal fin length. The other factors F4, F5 and F6 had a zero or negative value except for stoutness and Head length for F4 and F5 respectively, where the correlation was positive. It was further concluded that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 factors bi-plot and observation plot analysis.

In the PCA for the R. rita, the Eigen values showed that its trend reached its maximum at level of first factor. In the same way according to the Kaiser (1958) Criterion based upon the Eigen values greater than one, it can be assumed after observing the results that the PCA grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 99.02% of the cumulative variation among the morphometric parameters of study The first group accounted for 96.85% of the cumulative variability while the second major group accounted for 2.16% of the cumulative variability. The factor one was positively correlated for all studied parameters and factor two (F2) were negatively correlated with all the morphometric characters measured except for the adipose fin length for the R. rita. F3 was negatively correlated except for anal fin, F4 was positively correlated for dorsal fin length, anal fin length and adipose fin length, F5 had a positive correlation for head length dorsal fin length, adipose fin length while other parameters had a negative correlation ship, F6 had a positive correlation for head length and average length paired pectoral fin while negative for other observed parameters, F7 had a positive correlation for caudal fin and a negative for total length and average length paired pectoral fin and F8 were negatively correlated for body weight, fork length, stoutness and average length of the paired pelvic fins. The observation plot and bi-plot of the morphometric parameters further suggests that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on two major factors

364

The results obtained from the PCA indicated clearly that the augment in the number of factors or components was correlated with the decrease in Eigen values. In case of S. seenghala the values obtained from the PCA for the morphometric data showed that its trend reached its maximum at level of second factor. The results grouped the tested variables or parameters of the fish morphometery into two main components, which all together accounted for 99.91% of the cumulative variation among the morphometric parameters of study. The cumulative variability among two groups, the first major group accounted for 97.154% while the second accounted for 2.752%. The factor one (F1) was positively correlated when variables of morphometric characters. In case of factor F2 and F3, only the average length of the paired pelvic fins and average length of pectoral fins & the average length of the paired pelvic fins, respectively, were positively correlated, while the others were negatively correlated. In case of F4, the values of head length, stoutness, the average length of the paired pelvic fins and average length of paired pectoral fins were positively correlated. In case of F5, the values of head length, anal fin length, the average length of the paired pelvic fins and average length of pectoral fins were positively correlated. The others were negative in their relation. The relationship was positive in case of factor six (F 6) when parameters like stoutness, anal fin length, average length of paired pectoral fins and the average length of the paired pelvic fins were studied. In F7 head length, stoutness, caudal fin, anal fin length, average length of pectoral fins and the average length of the paired pelvic fins were positive while other parameters were negatively correlated. Total length was positive in factor F 8 in addition to positive parameters in factor F 7. In the F9 only the body weight and dorsal fin length were negative in correlation, while the others were positively correlated The observation plot and bi-plot of the parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups. On the observation of these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on observation plot analysis and the F1 & F2 factors bi-plot

The results obtained from the PCA according to the Kaiser (1958) Criterion based upon the Eigen values greater than one, first two main factors accounted for 99.996% of cumulative variation among the morphometric parameters of W. attu.

365

The first group amongst the major two groups accounted for 99.983% of the cumulative variability while the second accounted for 0.013% of the cumulative variability. The factor one (F1) was positively correlated with all the body parameters. Correlation was positive for head length only for F2 and head length and stoutness for F3, while others were negatively correlated. In factor four (F4), only the total length was negatively correlated and in factor five (F5) fork length, total length, head length and stoutness were positive in correlation.

The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups. On the observation of the these plots it was decided that the genotypes divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 factors bi-plot and observation plot analysis.

Morphometric comparisons of the African cat fish, Clarias gariepinus populations in Turkey was also conducted by TURAN et al., (2005) and the results are in accordance with the present studies. These results are also in accordance with the STERGIOU AND VASILIKI (2003) while working on length and girth interrelationship of the fishes of the marine environment, concluded that the said parameters of 18 marine species were significantly different statistically.

The results of present study are on the same line as the study of RASOOL et al., (2012a). In their studies on Morphometric Parameters in Hatchery Raised and Natural Populations of Labeo rohita, they found by that the body weight, total length and average length of paired pectoral fins of L. rohita were non-significantly different (P> 0.05), anal fin length was significantly different (P<0.05), and all the remaining parameters were highly significantly different (P<0.01) among the sites. The condition factor and length-weight relationship was also calculated. The correlation of fish body weight showed highly significant (p<0.001) and positive correlation with all the Morphometric parameters but the fork length of the L. rohita showed a positive and highly significant (p< 0.0001) correlation with all the parameters except with the caudal fin length where the correlation was also positive but non-significant.

366

The findings of present study are also in accordance with the work of RASOOL et al., (2012b) on the populations of Labeo rohita. In their study on morphometric parameters performed principle component analysis to determine the variation among the populations found that the PCA divided the fish populations of five locations on the basis of morphometery into two main components, which all together accounted for 80.27% of the cumulative variation among the morphometric parameters. The first group amongst the major two groups accounted for 64.245% of the cumulative variability while the second accounted for 16.028% of the cumulative variability. In same way the cumulative variation has been observed in all the fish species under study as mentioned above

This research represents an important advancement in understanding biology of commercial important carnivore fishes of the Indus Basin and highlights the morphological divergence in this widely distributed C. marulius and S. seenghala populations. Such divergence is commonly observed for freshwater and lacustrine populations (SCHLUTER AND MCPHAIL, 1992; SNORRASON et al., 1994). Present research has identified significant differences in the C. marulius and S. seenghala populations in the five locations studied. If shape is related to either environmental influences on larval development (CARDIN AND SILVA, 2005) or diversifying selection and ecological adaptation at a trophic level (COSTA AND CATAUDELLA, 2007), then spatially or latitudinally different environmental factors (e.g., temperature and resource availability) may explain the variations in morphometric heterogeneity of these species.

Physico Chemical parameters of the study sites

The data of the physico-chemical parameters from the study sites were analyzed by the Pearson correlation The correlation of the pH with water temperature (r = 0.107) and dissolved oxygen (r = 0.905) was positively non-significant while the correlation with electrical conductivity (r = -0.798), salinity (r = -0.888), total dissolved solids (r = -0.857), total alkalinity (r = -0.736) and total hardness (r = -0.499) was negatively non-significant. The correlation of the dissolved

367

oxygen with water temperature (r = 0.313) was positively non-significant while the correlation with electrical conductivity (r = -0.669), salinity (r = -0.828), total dissolved solids (r = -0.809), total alkalinity (r = -0.930) and total hardness (r = -0.300) was negative but also non-significant as like with the water temperature. The electrical conductivity was positively correlated with all the physic-chemical parameters as with water temperature (r = 0.482), salinity (r = 0.925), total dissolved solids (r = 0.889), total alkalinity (r = 0.452) and total hardness (r = 0.906) and this correlation was non- significant.

The salinity amongst the water parameters was correlated positively with water temperature (r = 193), total alkalinity (r = 0.717) and total hardness (r = 0.734) and it was non-significant but with total dissolved solids (r = 0.994) the correlation was also positive but highly significant (P <0.001). The total dissolved solids values observed from the study sites were positively correlated with water temperature (r = 0.172), total alkalinity (r = 0.734) and total hardness (r = 0.657) and this correlation was non-significant. The correlation between the total alkalinity and total hardness was also positive and non- significant (r = 0.048).

RAPD Data

The number of bands for PCR products of Channa punctatus by gel electrophoresis ranged from as low as three to a maximum of seven, with an average of 6 bands per primer. The number of polymorphic bands per primer was 1 to 3 amplified. The polymorphic bands in these populations ranged from 14.29% to 50%.

The bands of amplification products produced by primers by gel electrophoresis in Channa marulius ranged from as low as three to a maximum of seven, with an average of 6 bands per primer. 1 to 3 polymorphic bands per primer were obtained. The polymorphic bands in these populations ranged from 14.29% to 50%.

368

The number of bands by gel electrophoresis in the PCR product produced by primers in Rita rita ranged from as low as three to a maximum of seven, with an average of 6 bands per primer. 1 to 3 polymorphic bands per primer were observed. The polymorphic bands in these populations ranged from 14.29% to 50%.

The bands produced in amplified products by primers in PCR products by gel electrophoresis ranged from as low as three to a maximum of seven, with an average of 6 bands per primer in Sperata seenghala. 1 to 4 polymorphic bands per primer were found. The polymorphic bands in these populations ranged from 14.29% to 57.14%.

The number bands produced by gel electrophoresis ranged in Wallago attu from as low as three to a maximum of seven, with an average of 6 bands per primer in the amplified products. 1 to 4 polymorphic bands per primer were observed. The polymorphic bands in these populations ranged from 14.29% to 66.67%.

PCR products were studied and the data of the scorable bands of the all amplified primers obtained were subjected for the analysis for Jaccard’s coefficient by following the Unweighted Pair Group Method with Arithmetic Mean (UPGMA).

The dendrogram was developed which divided the randomly selected individuals of the five populations into four classes/clusters. The division of all the randomly selected five population representative C. punctatus samples collected from different geographical locations in the three cluster; 16 in first cluster/class, 4 in second cluster/class, 2 in the third and fourth class/cluster, each. The results of dendrogram of Channa punctatus, it is observed that majority of the samples are in one class and very few of them have got some genetic distinction

The dendrogram developed from the data of the scorable bands of all amplified primers divided the randomly selected individuals of the five populations of C. marulius into four classes/clusters as 21 samples in first cluster/class, one individual both in second and fourth cluster/class i.e., Cha3 collected from Chashma barrage and Tsa1 collected from Taunsa Barrage and two in the third class

369

All most all individuals of Channa marulius are grouped in one class and only four individuals form the remaining three classes. This means that these animals have got some distinction which may be due to the environmental changes. These results are in line with the CADRIN (2000) concluded from his experiment that it is often difficult to explain the causes of morphological differences between populations. These differences may be genetic differences, or they may be associated with phenotypic plasticity in response to different environmental factors in each area. These results showed that 84% population genetically same as making one class. Only the miner level of population is genetically different 16% samples divide into three classes. This means that a very miner change is there in the genetically makeup. So these individuals seem to be with distinct from the remaining populations of the same sites and support the hypothesis that the evolutionary process, environmental condition and other unknown factors are responsible for this distinctiveness.

The dendrogram formed by the analysis of the RAPD data for the randomly selected individuals of the five populations of R. rita into four classes/clusters. The division of the entire randomly selected five population representative into four classes was so that there were 21 individuals in first cluster/class, One individual in 2nd and 4th cluster/class, Cha3 collected from Chashma Barrage and Tsa1 collected from Taunsa barrage, respectively, while 2 individuals in the 3rd class/cluster.

The results of R. rita are confirmatory with the results of C. punctatus, C. marulius and S. seenghala as major group of the population is genetically same as grouped in one class and very few samples has got distinction.

The dendrogram developed by the presented data of the scorable bands of the all amplified primers in case of S. seenghala divided the randomly selected individuals of the five populations into four classes/clusters as follows; 20 in first cluster/class, 2 in second cluster/class and 3rd and 4th Class/cluster contains only one individual each, i.e., Trm1 collected from Trimu Barrage and Tsa1 collected from Taunsa Barrage, respectively.

370

Most of the samples collected in case of S. seenghala were grouped genetically into one class in cluster analysis. Although there was a highly significant difference in various morphological parameters when compared with different sampled locations subjected to ANOVA but this difference is not so depicted in this genetic analysis. This difference could be the result of environment gradient as reported by the LANGERHANS et al. (2003) found that the distance between habitats correlated positively with the level of divergence in body shape among conspecific populations of two Neotropical fish species. Although during this study the differences in morphology of fishes was observed, this variation was not supported by the RAPD data which may be explained by the fact that this morphological variation observed in this study may be the result of an environmental gradient rather than genetic isolation.

The dendrogram developed from the Principal Component Analysis (PCA) by the XLSTAT 2012 version 1.2 by the presented data of all the amplified primers divided the randomly selected individuals of five populations collected from different geographical locations of W. attu into six classes/clusters. i.e., 16 individuals in 1st class/cluster, 3 individuals in 2nd and 4th Class/cluster while the representative of classes/clusters 3rd, 5th and 6th were only one individual each. These results showed that 88% of all the populations were grouped into three classes while remaining three classes consist of only 12% i.e., 4% each. This shows that most of the populations of the W. attu were genetically similar with some distinction on minor level as Cha3 collected from the Chashma Barrage was the only individual in class three, Qbd5 collected from the Qadirabad Barrage was the only individual in class five while class six also consist of only one individual that was Blk4 collected from the Baloki Barrage. So these individuals seem to be with distinction from the remaining populations of the same sites and support the hypothesis that the evolutionary process and environmental condition and other unknown factors are responsible for this distinctiveness.

371

These results are also confirm the results of MOSTAFA et al., (2009) who suggested that minute difference in populations of Kalibasu were probably caused due to habitat degradation in many ways, which ultimately affects the genetic variation of Kalibasu.

The variance decomposition for C. punctatus for the optimum classification values was 73.83% for within class and 26.07% for the between class differences. The distance between the class/clusters centroids remained as 0.718 for class 1 and 2, 1.008 for class 1 and 3, 1.008 for class 1 and 4, 1.225 for class 2 and 3 and 2 and 4 while this distance was 1.414 for class 3 and 4. In this classification Cha2 from the Samples collected from Chashma Barrage, Cha5 collected from the same location, Qbd2 collected from Qadirabad Barrage and Qbd3 collected from the same location were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.414 between the central objects of class 1 and 2, 1.000 between the class 1 and 3, 4, respectively, 1.732 between the central objects of class 2 and 3, 4, respectively, while the distance between central objects of class 3 and 4 was 1.414.

The variance decomposition in C. marulius for the optimum classification values remained as 65.90% for within class variation and 34.10% for between class differences. The distance between the class/clusters centroids was as 1.737 for class 1 and 2, 1.105 for class 1 and 3, 1.009 for class 1 and 4, 1.500 for class 2 and 3, 2.000 for class 2 and 4, while this distance was 1.500 for class 3 and 4. In this classification Cha2 from the samples collected from Chashma Barrage, Cha3 collected from the same location, Qbd4 collected from Qadirabad Barrage and Tsa1 collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.732 between the central objects of class 1 and 2, 1.414 between the class 1 and 3, 1.000 between the central objects of class 1 and 4, 1.732 was the distance between central objects of class 2 and 3, 2.000 was

372

the distance between central objects of class 2 and 4, while this distance for central objects of class 3 and 4 was also 1.732.

The variance decomposition for the optimum classification values remained as 65.90% for within class variation and 34.10% for the between class differences in R. rita The distance between the class/clusters centroids remained as 1.737 for class 1 and 2, 1.105 for class 1 and 3, 1.009 for class 1 and 4, 1.500 for class 2 and 3, 2.000 for class 2 and 4, while this distance was 1.500 for class 3 and 4. In this classification Cha2 from the samples collected from Chashma Barrage, Cha3 collected from the same location, Qbd4 from the samples collected from Qadirabad Barrage and Tsa1 collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.732 between the central objects of class 1 and 2, 1.414 between the central objects of class 1 and 3, 1.000 between the central objects of class 1 and 4, 1.732 was the distance between central objects of class 2 and 3, 2.000 was the distance between central objects of class 2 and 4 while this distance for central objects of class 3 and 4 was 1.732.

The variance decomposition for S. seenghala the optimum classification values remained as 64.19% and 35.81% for within class and between class differences, respectively. The distance between the class/clusters centroids remained as 1.233 for class 1 and 2, 1.421 for class 1 and 3, 1.682 for class 1 and 4, 1.871 for class 2 and 3, 2.121 for class 2 and 4, while this distance was 2.236 for class 3 and 4. In this classification Cha2 from the Samples collected from Chashma Barrage, Qbd1 from the samples collected from Qadirabad Barrage and Trm1 collected from Trimu barrage at the junction of Chenab and Jhelum rivers and Tsa1 collected from Taunsa Barrage at River Indus were central objects for class/clusters 1, 2, 3 and 4, respectively. The distances between the central objects of the classes remained as; 1.414 between the central objects of class 1 and 2, 3 respectively, 1.732 between the central objects of class 1 and 4, 2.000

373

was the distance between central objects of class 2 and 3, while this distance for central objects of class 2 and 4 and 3 and 4 was 2.236, respectively.

The variance decomposition for W. attu, the optimum classification values remained as 49.14% for within class variation and 50.86% for the between class differences. The distance between the class/clusters centroids remained as 1.116 for class 1 and 2, 1.422 for class 1 and 3, 1.065 for class 1 and 4, 1.378 for class 1 and 5, 1.422 for class 1 and 6, 1.795 for class 2 and 3, 1.528 for class 2 and 4, 1.795 for class 2 and 5, 6; 1.764 for class 3 and 4, 2.000 for class 3 and 5, 6; 1.764 for class 4 and 5, 6 while this distance was 2.000 for class 5 and 6. In this classification Cha5 from the Samples collected from Chashma Barrage, Blk2 collected from the Baloki Barrage At River Ravi, Cha3 from the Samples collected from Chashma Barrage, Qbd1 and Qbd5 from the samples collected from Qadirabad Barrage, while Blk4 collected from the Baloki Barrage at River Ravi were central objects for class/clusters 1, 2, 3, 4, 5, and 6, respectively.

The results of present study are in accordance with the results of CHAUHAN et al., (2007) who studied different populations of wild C. mirgala from different River basins and concluded that there existed low level of differentiation between the populations of the same species and this may be due to common ancestry and exchange of individuals among the River basins. The results are also confirmation of the results indicated in the study conducted by DAYU et al., (2007) on the genetic similarity amongst the wild populations of Cyprinus carpio. They concluded that there was a correlation between the clustering result and the geographical distribution.

These results are also comparable with the results of MOHINDRA et al., (2007) on the genetic variability in three clariid species, Clarias batrachus, C. gariepinus and C. macrocephalus and the UPGMA phylogenetic tree revealed three distinct clusters: C. batrachus; C. gariepinus and C. macrocephalus.

The results of our study are completely complementary to the studies of RASOOL et al., (2012c). In their study on Indian major carp, C. mirgala performed Clustering Analysis for Intraspecific Variation amongst the Populations of the fish and 374

reported that data of the morphometric parameters divided the populations of in to four major clusters or classes. They further reported that variance decomposition for the optimal classification values remained as, 27.28% for within class variation while 72.72% for the between class differences. The distance between the class/cluster centroids remained as; 50.820 for class one and two, 18.063 for class one and three, 14.564 for class one and four, 68.856 for class two and three, 36.708 for two and four while this distance between class three and four centroids was 32.408.

These results are comparable with ZHU et al., (2009) who concluded that the reason for morphological diversity within the sub populations was as result from the variable environments.

The results obtained from the PCA regarding cumulative variability for C. punctatus indicated, first four main factors accounted for 55.849%. The first and second group (F1 and F2) accounted for 14.881% each, while the third and fourth (F3 and F4) 14.069% and 12.018%, respectively, of the cumulative variability. The values are 7.764%, 7.605% for fifth and sixth factor, respectively, 7.440% for seventh, eighth and ninth, each, the values are 4.310% and 2.151% for F10 and F11, respectively. The trends of variables selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations. This trend divided the role of primers into four major variable groups, one group towards the positive side and one group towards the negative side and two groups had the neutral affect.

After the varimax rotation according to the Kaiser (1958) criterion, the cumulative variance is same as before rotation as 55.849% with some variation in individual factor variation which remained as 14.881% and 14.881% for D1 and D2, each.

The trend of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into four major variable groups, one group towards the positive side, one towards negative and two

375

groups at the neutral site. The individuals of five populations taken as observations and divided into two major factors after varimax rotation indicated that the representative individuals of five populations are genetically correlated with each other with some exceptional case like Cha5 sample from Chashma Barrage towards somewhat negative value and Blk5 sample from the Baloki Barrage of the River Ravi towards the extreme negative distinction. The Trm2 is at extreme positive distinction. This indicates that the some environmental impacts are showing their influence towards the genetic drifts. The critical Chi- square value is 32.671 for Channa punctatus.

The PCA indicated clearly that the values were at its maximum at the level of first factor. According to the Kaiser (1958) Criterion based upon the Eigen values greater than one, first three main factors accounted for 42.770% of cumulative variability in C. marulius. The PCA grouped the tested variables or parameters of the fish RAPAD amplification data into three main components which all together accounted for 42.78% of the cumulative variation among the factors. The first, second and third group (F1, F2 and F3) amongst the major three groups accounted for 16.162% 14.415% 12.193% each, of the cumulative variability while the fourth to ninth (F4 to F9) from these accounted for 8.013% each and tenth and eleventh (F10 and F11) 5.091 and 4.062% respectively, of the cumulative variability. The representative individuals of five populations on the basis of two major factors were accumulated as one group.

According to the Kaiser (1958) criterion, after varimax rotation in C. marulius, the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2 here, the cumulative variance was 30.577% with some variation in individual factor variation which remained as 15.580% and 14.997% for D1 and D2, respectively. The trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into five major variable groups, two groups towards the positive side and two groups towards the negative side with one at neutral place. The representative individuals of five populations are genetically positively correlated with each other with some

376

exceptional cases like Cha3 sample from the Chashma Barrage of the River Indus towards the extreme negative value and Tsa1, Qbd2 and Trm1 at extreme positive site. This indicates that the some environmental impacts are showing their influence towards the genetic drifts.

The PCA grouped the tested variables or parameters of the Channa marulius into two main components, which together accounted for 99.996% of the cumulative variation among the morphometric parameters of study Factor three (F3) has a value of 0.003% only. The factors one (F1) had a positive correlation for all the factors whereas the factor two (F2) were negatively correlated with all the parameters except dorsal fin length. The other factors F4, F5 and F6 had a zero or negative value except for stoutness and head length for F4 and F5 respectively, where the correlation is positive The observation plot and bi-plot of the morphometric parameters were drawn by using the first two factors on the basis of which the parameters were grouped into two major groups In the Bartlett's sphericity test for C. marulius The critical Chi- square value is 41.337. Squared cosine values were largest for all different parameters in factor F1 from all the places of study. On the observation of these plots it was decided that the genotypes that were divided into five classes by the cluster analysis were also in the same groups based on the F1 and F2 observation plot and factors bi-plot analysis.

The results obtained from the PCA of R. rita indicated clearly that the increase in the number of factors or components is correlated with the decrease in Eigen values. The values were at its maximum in the level of first and second factor. According to the Kaiser (1958) Criterion based upon the Eigen values greater than one, the PCA grouped the tested or parameters of the fish RAPD amplification data into two main components which all together accounted for 30.97% of the cumulative variation among the factors. The first and second group (F1 and F2) amongst the major two groups accounted for 20.609% and 10.359%, respectively, of the cumulative variability while the third to ninth group (F3 to F9) from these accounted for 8.681% each respectively, of the cumulative variability. The trend of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected

377

individuals of five populations divided the role of primers into five major variable groups, three groups towards the positive side and one group towards the negative side while the remaining at the neutral distinction. In the figure-61, it is clearly indicated that the representative individuals of five populations on the basis of two major factors are accumulated as a single group.

After the varimax rotation was performed to show the influence or share of each individual. According to the Kaiser (1958) criterion in R. rita, the cumulative variance was 30.968% with some variation in individual factor variation which remained as 20.141% and 10.827% for D1 and D2, respectively.

The trends of most variable selected two factors after varimax rotation D1 and D2 towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into five major variable groups, two groups towards the positive side and one group towards the negative side and the two remaining at the neutral place. The representative individuals of five populations were genetically positively correlated with each other with some exceptional cases like Tsa2 sample from Taunsa Barrage towards somewhat negative value and Cha3 sample from the Chashma Barrage of the River Indus towards the extreme negative value. On the other hand, the sample Qbd4 and Cha1 was at extreme positive limit. This indicates that the some environmental impacts are showing their influence towards the genetic drifts.

The PCA results for S. seenghala show that the increase in the number of factors or components was correlated with the decrease in Eigen values and were at their maximum at the level of first and second factor. According to the Kaiser (1958) Criterion based upon the Eigen values greater than one, first four main factors accounted for 55.406% of cumulative variability. The PCA grouped the tested variables or parameters of the fish RAPD amplification data into four main components. The values for individual factors were 15.717%, 14.044%, 13.889%, 11.756% for first, second, third and

378

fourth group (F1, F2, F3 and F4) respectively. The factors fifth to ninth accounted for 6,944% each. The factors tenth, eleventh and twelfth (F10, F11 and F12) accounted for 4.815%, 3.623% and 1.433% respectively.

The trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into seven major variable groups’ two groups towards the positive side and four groups towards the negative side and one has neutral distinction. The representative individuals of five populations on the basis of two major factors are accumulated as a one group.

The varimax rotation that alters the axes in the PCA increases the cumulative variability of the squared loading. According to the Kaiser (1958) criterion, the percentage of variability where the first two most common factors F1 and F2 has been rotated orthogonally and designated as D1 and D2, the cumulative variance was 29.761% with some variation in individual factor variation which remained as 15.694% and 14.067% for D1 and D2, respectively.

The trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into seven major variable groups, two groups towards the positive side and four groups towards the negative side and one at the neutral distinction. The individuals of five populations taken as observations were divided into two major factors after varimax rotation. The representative individuals of five populations are genetically positively correlated with each other with some exceptional cases like Qbd3 and Blk3 with somewhat negative value and Tsa1 with extreme negative value along with Qbd1, Tsa2, trm1 and Cha3 at extreme positive places. This indicates that the some environmental impacts are showing their influence towards the genetic drifts.

379

The results obtained from the PCA for W. attu clearly indicated that the increase in the number of factors or components was correlated with the decrease in Eigen values. The values according to the Kaiser (1958) Criterion based upon the Eigen values greater than one, the tested parameters of the fish RAPD amplification data into four main components which all together accounted for 50.37% of the cumulative variation among the factors. The first and second group (F1 and F2) accounted for 13,043% each, of the cumulative variability while the third and fourth (F3 and F4) from these accounted for 12.764% and 11.544% respectively, of the cumulative variability. The fifth factor (F5) accounted for 9.154% while the factors sixth to tenth (F6 to F10) from these accounted for 6.522% each. Factor eleventh, twelfth and thirteenth (F11, F12 and F13) has a value of 4.043%, 2.588% and 1.211% respectively, of the cumulative variability.

The trends of most variable selected two factor in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into four major variable groups, one group towards the positive side, one groups toward the negative side and two groups has the neutral distinction. The representative individuals of five populations on the basis of two major factors were accumulated as a one group.

According to the Kaiser (1958) criterion, the varimax rotation criterion maximizes the sum of the variances of the squared coefficients within each eigenvector, and the rotated axes remain orthogonal. F1 and F2 rotated orthogonally and designated as D1 and D2 and the cumulative variance was 26.087% with some variation in individual factor variation which remained as 13.043%, each.

The trends of most variable selected two factors after varimax rotation D1 and D2 in which the variables are the different primers towards their contribution for polymorphism amongst the randomly selected individuals of five populations divided the role of primers into four major variable groups, one group towards the positive side and one group towards the negative side and the two with neutral affect. The individuals of five populations taken as observations and divided into

380

four major factors after varimax rotation, are genetically correlated with each other with some exceptional case like Qbd5 with some positive correlation and Cha3 with extreme positive value and Blk4 sample from the Baloki Barrage of the River Ravi towards the extreme negative value. This indicates that the some environmental impacts are showing their influence towards the genetic drifts.

The results of the present study showed that most of the individuals of five populations collected from different geographical locations were grouped into same class/cluster, which is the indication that the parental stock of all individuals of the same species were same but minute levels of differentiation are due to some evolutionary process or environmental impacts. These results are in accordance with the studies conducted by SAINI et al., (2010) examined the genetic variability between the populations of S. seenghala in river Sutlej and river Beas of Indus river system in India and found following 95% criterion as standard and found that there was 89.06% for Beas population as compared to 95.31% for Sutlej population, they found a moderate level of genetic divergence, 0.0486 between both the populations which is the result of substructure of the S. seenghala in both the rivers.

These results are also in line with findings of DANISH, et al. (2012) on the studies of Molecular characterization of two populations of catfish Clarias batrachus L. using random amplified polymorphic DNA (RAPD) markers and found that the similarity within the population from wild varied from 0.40 to 0.83 with a mean ± SE of 0.57 ± 0.08. The Jaccard’s similarity coefficient ranged from 0 to 0.27. At 0.06 similarity coefficient, two major clusters were formed, which indicates that the genotypes belonging to same clusters were genetically similar and those belonging to different clusters were dissimilar. Significant (P < 0.05) population differentiation indicated some degree of intra- and inters- population genetic variations in two populations of catfish. In the present study the difference is very minor between the samples collected from different sampling sites which showed that these fish belong to same genetic group.

381

The results of the present study are in accordance with the studies of MATOSO et al., (2012) when studied the Steindachneridion melanodermatum in sympatry caught in nature from Lguacu River and genetic variability of wild parents and F1 fish born in captivity. The analysis of specimens with regard to patterns of RAPD molecular markers showed genetic similarity ranging from 0.57 to 0.95; two groups were determined for the wild specimens. The results suggested different genetic lineages in sympatry in nature. Heterozygosity and percentage of polymorphic loci were 0.31 and 79% and 0.23 and 62%, respectively, for the two populations of wild specimens and 0.26 and 66%, respectively, for those born in captivity. These results are according to the findings of LEUZZI et al., (2004) in their study by the RAPD technique when they analyzed the genetic structure of populations of the fish Astyanax altiparanae (Characidae, Tetragonopterinae) living in the lower, middle and upper Paranapanema River, Brazil. They reported that genetic variability to be 42.64%, 75% and 75% in the low, middle and upper reaches, respectively. The dendrogram of genetic similarity, obtained by comparative analysis of the sets of samples from the three sites, showed the formation of three clusters. All of the genetic parameters used indicate that the population in the lower Paranapanema is genetically different from those in the middle and upper sections. The theta P test shows that the low Paranapanema is highly differentiated from the middle (0.2813) and upper (0.2912) Paranapanema, while the differentiation between the last two is moderate (0.0895).

ALMEIDA, et al., (2003) concluded such results while working with the Pimelodus maculatus populations from the Tietê and Paranapanema rivers in Brazil on genetic structure analyzed by using RAPD markers. The proportion of polymorphic loci was greater than 50% in the populations of both rivers. Genetic diversity data showed that, in spite of its nine hydroelectric plants, the Tietê river population was genetically homogeneous, whereas the Paranapanema river population was structured. This might be due to the presence of high waterfalls distributed all along its course.

382

NAGARAJAN et al., (2006) while working on C. punctatus populations for genetic variations collected from three rivers of south India with randomly amplified polymorphic DNA (RAPD) postulated somewhat same results as like the present study. Among the three populations, the highest genetic identity (0.9231) was found between Thirunelveli and Quilon populations. The results of the present study demonstrated that Thirunelveli and Quilon populations are more related to each other than to the Coimbatore population. MUNEER et al., (2012) postulated same results as like the present study in their the study on endemic and endangered yellow catfish (Horabagrus brachysoma) sampled from three locations in Western Ghats river systems of India to find out the comparative assessment of genetic diversity using allozymes, random amplified polymorphic DNA (RAPD), and micro satellite markers. Among the three markers, micro satellites show more polymorphism, having 100% polymorphic loci, whereas allozymes show the least (56%). In RAPD, 60.5% of fragments were polymorphic. Observed heterozygosity and F ST values were very high in micro satellites, compared with the other markers. Micro satellite and RAPD markers reported a higher degree of genetic differentiation than allozymes among the populations depicted by pair wise F ST/G ST, AMOVA, Nei’s genetic distance, and UPGMA dendrogram. The three classes of markers demonstrated striking genetic differentiation between pairs of H. brachysoma populations. The data emphasize the need for fisheries management, conservation, and rehabilitation of this species.

The results of the present study are coherent to the TAMANNA et al., (2012) studied the genetic variation in three wild populations of the striped dwarf catfish, Mystus vittatus (Bloch) namely Chalan beel (Natore), Mohangan jhaor (Netrakona) and Kangsha river (Netrakona) in Bangladesh and postulated somewhat same results as in the present study. The proportions of polymorphic loci were more or less similar in the three studied populations. The intra-population similarity index, gene diversity and Shannon’s Information Index were found to be the highest in the Chalan beel population followed by those of the Kangsha river and Mohangan jhaor population. The inter population similarity index was highest between Chalan beel and Kangsha river population and the gene flow was highest between Mohangan jhaor and Kangsha river population. The population differentiation values (PhiPT) were found to be insignificant indicating no

383

significant differentiation among the three populations. No population-specific bands were detected. The RAPD analysis revealed a high level of genetic variation in the three wild populations of the dwarf catfish M. vittatus.

Knowledge of genetic structure of the major River populations is helpful for management of the populations in order to maintain their genetic quality. In this study the results indicate good correspondence in the data analyses of morphometric parameters, and RAPD molecular markers using various statistical techniques with the exception of the distinction of one individual, Tsa1 collecte dfrom Taunsa Barrage and to some extent specimens from Qadirabad barrage as well, which clearly indicated some environmental impacts, are likely influencing the genetic makeup within and between the local populations.

384

Chapter 6

REFERENCES

AGNESE, J.F., TEUGELS, G.G., GALBUSERA, P., GUYOMARD, R. AND VOLCKAERT F., 1997. Morphometric and genetic characterization of sympatric populations of Clarias gariepinus and C. angullaris from Senegal. J of Fish Biol, 50, 1143–1157.

385

AGRICULTURAL SCIENCE, 14th June, 2012. Molecular Phylogenetics and population genetic structure of genus Pelteobagrus (Teleostei, Bagridae). Available from Agriculuture Science Papers. www.agrpaper.com [Accessed 12 July 2012].

AHMAD, M., 1943. Fauna of Lahore, 5. Fishes of Lahore. Bull Dep Zool Univ Punjab, 1, 253-37 4.

AHMAD, N., 1960. Fauna of Lahore. 5. Fishes of Lahore. Bull Dep Zool Univ Punjab, 1, 253-374.

AHMAD, N., 1963. Freshwater fish fauna of West Pakistan. Agri. Pakistan, 14, 77-82.

AKHTAR, N., 2001. Strategic Planning for inland fisheries and aquaculture sector in Pakistan. In: Proceedings of National seminar on strategic planning for Fisheries and Aquaculture to face the challenges of new Millennium. Karachi- Pakistan.

ALI, A.B., 1999. Aspects of the reproductive biology of female snakehead (Channa striata Bloch) obtained from irrigated rice agro ecosystem, Malaysia. Hydrobiologia, 411, 71-77.

ALI, B.A., HUANG, T-H., QIN, D-N.AND WANG X-M., 2005. A review of rom amplified polymorphic DNA (RAPD) markers in fish research. Rev Fish Biol Fish, 14, 443–453.

ALJANABI, S.M. AND MARTINEZ, I., 1997. Universal and Rapid salt-extraction of high quality genomic DNA for PCR- based techniques. Nucl Acids Res, 25, 4692-4693.

ALMEIDA, F.S., SODRE, L.M.K. AND CONTEL, E.P.B., 2003. Population Structure Analysis of Pimelodus maculates (Pisces, Siluriformes) from the Tiete and Paranapanema Rivers (Brazil). Genet Mol Biol, 26, 301–305.

386

AMBAK, M.A., BOLONG, A.A., ISMAIL, P. AND TAM, B.M., 2006 Genetic variation of snakehead fish (Channa striata) populations using random amplified polymorphic DNA. Biotechniques. 5, 104–110.

AMERICAN PUBLIC HEALTH ASSOCIATION, 1998. Standard methods for the examination of water and wastewater. 20th. Ed., Washington DC., USA. 1220.

AMILHAT, E. AND LORENZEN, K., 2005. Habitat use, migration pattern and population dynamics of chevron snakehead Channa striata in a rain fed rice farming landscape. J Fish Biol, 67 (Supplement B), 23-34.

APOSTOLIDIS, A.P., GELIA, D. AND MAMURIZ, Z., 2011. Genetic diversity among Balkan trout populations based on RAPD analysis. Russ J Genet, 47(8), 973-978.

APPLEYARD, S.A., GREWE, P.M., INNES, B.H. AND WARD, R.D., 2001. Population structure of yellow fin tuna (Thunnus albacores) in the western pacific ocean inferred from microsatellite loci. Mar Biol, 139, 383–393.

ARUL, V., 2008. Effects of delayed feeding on growth and survival of Channa striatus (Bloch) larvae. Aquac Res, 22, 423- 434.

AVSAR, D., 1994. A stock differentiation study of the sprat (Sprattus sprattus phalericus Risso) off the southern coast of the Black Sea. Fish Res, 19, 363–78.

BAGLEY, M.J., ANDERSON, S.L. AND MAY, B., 2001. Choice of methodology for assessing genetic impacts of environmental stressors polymorphism and reproducibility of RAPD and AFLP fingerprints. Ecotoxicology, 10, 239– 244.

BAIE, S.H. AND SHEIKH, K.A., 2000. The wound healing properties of Channa striatus- cetrimide cream–tensile strength measurement. J Ethno pharmacol, 71(1-2), 93-100.

387

BANARESCU, P. AND NALBANT, T.T., 1966. The 3rd Danish Expedition to Central Asia, Zoological results-34. Cobitidae (Pisces) from Afghanistan and Iran. Vid Medd Dansk Natural Foren, 129, 149-186.

BARDACH, J.E., RYTHER, J.H. AND MCLARNEY, W.O., 1972. Japanese yellowtail culture. Aquaculture, The Farming and Husbandry of Freshwater and Marine Organisms. Wiley-Interscience, New York. 868.

BARMAN, H.K., BARAT, A., BHARAT, M., YADAV, BANERJEE, S., MEHER, P.K, KRISHNA, REDDY, P.V.G. AND JANA, R.K., 2002. Genetic variation between four species of Indian major revealed by random amplified polymorphic DNA assay. Aquaculture, 217, 115–123.

BARMAN, H.K., BARAT, A., YADAV, B.M., BANERJEE, S., MEHER, P.K., REDDY, BARTFAI, R., EGEDI, S., YUE, G.H., KOVACS, B., URBANYI, B., TAMAS, G., HORVATH, L. AND ORBAN, L., 2003. Genetic analysis of two common carp brood stocks by RAPD and microsatellite markers. Aquaculture, 219, 157–67.

BARTISH, I.V., GARKAVA, L.P., RUMPUNEN, K. AND NYBOM, H., 2000. Phylogenetic relationships and differentiation among and within populations of Chaenomeles lindl. (Rosaceae) estimated with RAPDs and isozymes. Theor Appl Genet, 101, 554–561.

BASAVARAGA, N., MAHESH, T., GIRESSHA, O., MALLESHAPPA, H.D. AND VARGHESE, T.G., 1999. Carp seed production in Karnataka (India). Aquacul Asia (NACA), 4(1), 16-20.

BEACHAM, T.D., MCINTOSH, B. AND MAC-CONNACHIE, C., 2002. Microsatellite identification of individual sockeye salmon in Barkley Sound British Columbia. J Fish Biol, 61, 1021–1032.

BEAUMONT, A.R. AND HOARE, K., 2003. Biotechnology and genetics in fisheries and aquaculture. Blackwell Science, Oxford, 158 pp.

388

BEGG, G.A., FRIEDLAND, K.D. AND PEARCE, J.B., 1999. Stock identification and its role in stock assessment and fisheries management: n overview. Fish Res, 43, 1–8.

BERRA, T.M., 2001. Freshwater fish distribution. Academic Press, San Diego. 606

BHARDWAJ, R.M., 2005. Water quality monitoring in India - Achievements and Constraints. Inter-secretariat Working Group on Environment Statistics (IWG-Env), International Work Session on Water Statistics, 20–22 June 2005, Vienna, Austria.

BHAT, A.A., HANIFFA, M.A., DIVYA, P.R., GOPALAKRISHNAN, A., MILTON, M.J., KUMAR, R. AND PARAY, B.A., 2012. Molecular characterization of eight Indian Snakehead species (Pisces: Perciformes Channidae) using RAPD markers. Mol Biol Rep, 39, 4267–4273.

BHATTI, M.Z., MIRZA, Z.S., AND QURESHI, A.H., 1996. Status of fish seed production in Punjab. Book of Abstracts of Second Symposium on Fish and Fisheries of Pakistan held at Zoology Dept, Government College Lahore, Pakistan on 25-26 (Abstract).

BLOCH, M.E. AND SCHNEIDER, J.G., 1801. M.E. Blochii, Systema Ichthyologiae iconibus cx illustratum Post obitum auctoris opus inchoatum absolvit, corexit, interpolavit Jo. Gottlob Schneider, Saxo berolini. Sumtibus Austoris Impressum et bibliopolio Sanderiano Commissum. Systema Ichthyol, 1-584.

BLOCH, M.E., 1794. Nat. Ausland. Fische. 7, p. 141

BONOU, C.A. AND TEUGELS, G.G., 1985. Re ´vision syste ´matique du genre Parachanna Teugelset Daget, 1984 (Pisces: Channidae). Rev Hydrobiol Trop, 18, 267–280.

389

BOWEN, B.W., MEYLAN, A.B. AND PERRAN, R., 1992. Global population structure and natural history of Green Turtle (Chelonia mydas) in terms of matriarchal phylogeny. Evolution, 46, 865–881

BRZESKI, V. J., AND DOYLE, R. W. (1988). A morphometric criterian for sex discrimination in Tilapia. In: R. S. V. Pullin, T. Bhukaswan, K. Tonguthai & J. L. Maclean (Eds.), The Second International Symposium on Tilapia in Aquaculture (Vol. 15). 439-444: Manila: ICLRAM Conference Proceedings.

BUTH, D.G. and CRABTREE, C.B., 1982. Genetic variability and population structure of Catostomus santaanae in the Santa Clara drainage. Copeia, 2, 439-44.

BUTT, J.A. AND BUTT, A. A., 1988. An addition to the Fishes of Azad Kashmir. Scient Khyb, 1, 77-84.

CACCONE, A., ALLEGRUCCI, G., FORTUNATO, C. AND SBORDONI, V., 1997. Genetic differentiation within the European sea bass D. labraxas revealed by RAPD-PCR assays. J Hered, 88, 316–324.

CADRIN, S.X., 2000. Advances in morphometric identification of fishery stocks. Rev Fish Biol Fish, 10, 91–112.

CALLEJAS, C., AND OCHANDO, M. D., 2002. Phylogenetic relationships among Spanish Barbus species (Pisces, Cyprinidae) shown by RAPD markers. Heredity, 89(1), 36-43.

CARDIN, S. X., AND SILVA, V. M., 2005. Morphometric variation of yellowtail flounder. ICES Journal of Marine Science, 62, 683-694.

CARVALHO, G.R., 1993. Evolutionary aspects of fish distribution: genetic variability and adaptation. J Fish Biol, 43, 53– 73.

CHANDRA, S. AND BANERJEE, T.K., 2004. Histopathological analysis of the respiratory organs of Channa striata subjected to air exposure. Veterinarski Archi, 74, 37-52.

390

CHONDAR, S.L., 1999. In Biology of fin fishes and shellfishes. SCSC Publishers (India), Howrah, West Bengal, India. 514.

CHONG, L.K., TAN, S.G., YUSOFF, K. AND SIRAJ, S.S., 2000. Identification and characterization of Malaysian river catfish, Mystus nemurus (C and V) RAPD and AFLP analysis. Bioche Genet, 38 (3, 4), 63–76.

COSTA, C. AND CATAUDELLA, S., 2007. Relationship between shape and trophic ecology of selected species of sparids of the Caprolace coastal lagoon (central Tyrrhenian Sea). Environmental Biology of Fishes, 78, 115-123.

COURTENAY, W.R. AND WILLIAMS, J.D., 2004. Snakeheads (Pisces, Chan- nidae) a biological synopsis and risk assessment. USGS Circular, 1251, 143.

CURZ, E.M. AND LAUDENICA, I.L., 1980. Polyculture of milk fish (Chanos chanos Forskal) all male Nile tilapia (Tilapia nilotica) and snakehead (Ophiocephalus striatus) in fresh water ponds with supplemental feeding. Aquaculture, (20), 231-237.

DANISH, M., SINGH, I.J., GIRI, P. AND SINGH, C. P., 2012. Molecular characterization of two populations of catfish Clarias batrachus L. using random amplified polymorphic DNA (RAPD) markers. Afr J Biotechnol, 11(77), 14217- 14226.

DARS, B.A., NAREJO, N.T. AND AWAN, K.P., 2012. Morphometric, meristic characters and their relationships in Channa punctatus (Bloch) from River Indus near Jamshoro Sindh, Pakistan. Sindh Univ Res Jour (Sci. Ser.), 44 (1), 91-96.

DELLING, B., CRIVELLI, A.J., RUBIN, J-F. AND BERREBI, P., 2000. Morphological variation in hybrids between Salmo marmoratus and alien Salmo species in the Volarja stream, Soca river basin, Slovenia. J of Fish Biol, 57, 1199– 1212.

391

DHILLON, S.S. AND KAUR, H., 1996. Analytical studies on the aquatic ecosystems of Punjab. Final Technical Report, Punjab State Council for Science and Technology, Chandigarh.

DINESH, K.R., LIM, T.M., CHUA, K.L., CHAN, W.K. AND PHANG, V.P.E., 1993. RAPD analysis an efficient method of DNA finger-printing in fishes. Zool Sci, 10, 849–54.

DINESH, K.R., LIM, T.M., CHUA, K.L., CHAN, W.K. AND PHANG, V.P.E., 1996. Genetic variation inferred from RAPD fingerprinting in three species of tilapia. Aquaculture Int Dordrecht, 4, 19–30.

EHRLICH, P.R. AND WILSON, E.O., 1991. Biodiversity studies: Science and policy. Science, 253, 758-762.

FERGUSON, A., TAGGART, J.B., PRODÖHL, P.A., MCMEEL, O., THOMPSON, C., STONE, C., MCGINNITY, P. AND HYNES, R.A., 1995. The application of molecular markers to the study and conservation of fish populations, with special reference to Salmo. J Fish Biol, 47, 103-126.

FOOD AND AGRICULTURE ORGANIZATION OF UNITED STATES, 2010. Fisheries and aquaculture country profile- Pakistan. http //www.fao.org

FOOD AND AGRICULTURE ORGANIZATION OF UNITED STATES, 2012. Consumption of Fish and Fishery Products. Retrieved October 2, 2012, from http://www.fao.org/fishery/statistics/global-consumption/en.

FROESE R. AND PAULY D., 2012. Fish Base (version Jan 2012). In: Species 2000 and ITIS Catalogue of Life, 25th June 2012 (BISBY, F., ROSKOV, Y., CULHAM, A., ORRELL, T., NICOLSON, D., PAGLINAWAN, L., BAILLY N., KIRK, P., BOURGOIN, T. AND BAILLARGEON, G., eds). Digital resource at www.catalogue of life.org/col/. Species 2000 Reading, UK.

392

GARG, R., SILAWAT, N., SAIRKAR, P., VIJAY, N., AND MEHROTRA, N., 2010. RAPD analysis for genetic diversity of two populations of Mystus vittatus (Bloch) of Madhya Pradesh, India. Afr J Biotechnol, 8(17), 4032-4038.

GARG, R.K. AND SHARMA R., 2012. Molecular analysis of genetic diversity and in-situ approaches for conservation of a freshwater catfish Sperata seenghala (SYKES, 1839). Proc. 99th Indian Science Congress, Part II: Presidential Address Section II: Animal, Veterinary and Fishery Sciences 3-7 January 2012 Retrieved October 2, 2012, from www.isc2012.com/pdf/proceedings/animalveterinaryfisherysc.pdf

GIRI, P., TAJ, G. AND GINWAL, H.S., 2012. Molecular characterization of six populations of Acorus calamus L. using randomly amplified polymorphic DNA (RAPD) markers. Afr J Biotechnol, 11(40), 9522-9526. Available online at http //www.academicjournals.org/AJB. [Accessed on July 17, 2012].

GOPALAKRISHNAN, A., MUSAMMILU, K.K., MUNEER, P.M.A., LAL, K.K., KAPOOR, D., PONNIAH, A.G. AND MOHINDRA, V., 2004. Microsatellite DNA markers to assess population structure of red tailed barb Gonoproktopterus curmuca. Current Zoology (formerly Acta Zoologica Sinica), 50(4), 686-690.

GOVERNMENT OF PAKISTAN, 2012. Economic Survey of Pakistan (2011-12). Islamabad, Pakistan: Ministry of Finance, Government of Pakistan.

GUNTHER, A., 1864. Catalogue of fishes in the British Museum, London, volume i-v, 1-455.

HABIB, M., LAKRA, W.S., KHARE, V.M.P., BARMAN, A.S., SINGH, A., LAL, K.K., PUNIA, P. AND KHAN, A.A., 2011. Evaluation of Cytochrome b mtDNA sequences in genetic diversity studies of Channa marulius (Channidae: Perciformes). Mol Biol Rep, 38(2), 841–846.

HAMILTON, F., 1822. An account of fishes found in the River Ganges and its tributaries, Edinburgh and London. i-iv, 1- 405.

393

HANIFFA, M.A, MERLIN, T. AND MOHAMED, J.S., 2000. Induced spawning of the striped murrel Channa striatus using pituitary extracts, human chorionic gonadotropin, luteinizing hormone releasing hormone analogue, and ovaprim. ActaIcht Piscat, 30, 53-60.

HANIFFA, M.A., NAGARAJAN, M., GOPALAKRISHNAN, A. AND MUSAMMILU, K.K., 2007. Allozyme variation in threatened freshwater fish spotted Murrel (Channa punctatus) in south Indian River system. Biochem Genet, 45(3/4), 363–373.

HOSSAIN, M.K, LATIFA, G.A. AND RAHMAN, M.M., 2008. Observations on induced breeding of snakehead murrel, Channa striatus (Bloch, 1793). Int J Sustain Crop Prod, 3, 65-68.

HUANG, C-F., LIN, Y-H., AND CHEN, J-D., 2005. The Use of RAPD Markers to Assess Catfish Hybridization. Biodivers and Conserv, 14(12), 3003-3014.

HUSSAIN, Z., 1973. Fish and fisheries of the lower Indus basin (1966-67), Agri. Pakistan, 24, 170-188.

IBANEZ-AGUIRRE, A.L., CABRAL-SOLIS, E., GALLARDO-CABELLO, M. AND ESPINO-BARR, E., 2006. Comparative morphometrics of two populations of Mugil curema (Pisces: Mugilidae) on the Atlantic and Mexican Pacific coasts. Scientia Marina, 70 (1), 139–45.

IRRIGATION SYSTEM MAP. Hydropower potential list Punjab. www.irrigation.gov.pk. Retrived on September 2012.

JACCARD, P., 1901. Étude comparative de la distribution florale dans une portion des alpes et des jura. Bull Soc Vaudoise Sci Nat, 37, 547-579.

394

JAMALUDDIN, J.A.F., PAU, T.M. AND SITI-AZIZAH, M.N., 2011. Genetic structure of the snakehead murrel, Channa striata (channidae) based on the Cytochrome c oxidase subunit I gene: Influence of historical and geomorphological factors. Genet Mol Biol, 34(1), 152–160.

JAYARAM, K.C., 1971. Contributions to the study of Bagrid fishes generic status of Aorichthys Wu (Siluroidae: Bagridae). Proc Zool Soc Calcutta. 24(2), 149–156.

JOHNS, G.C. AND AVISE, J.C., 1998. A comparative summary of genetic distances in the vertebrates from the mitochondrial cytochrome b gene. Molecular Biology and Evolution, 15, 1481-1490.

JOHNSON, S.L., MIDSON, C.N., BALLINGER, E.W. AND POSTLETHWAIT, J.H., 1994. Identification of RAPD primers that reveal extensive polymorphisms between laboratory strains of zebra fish. Genomics, 19, 152–156.

KAWARAZUKA, N. (2010). The contribution of fish intake, aquaculture, and small-scale fisheries to improving nutrition: A literature review. The World Fish Center Working Paper No. 2106: The World Fish Center, Penang, Malaysia.

KHALEDI, H., REZVANIGILKOLAEI, S., ZOLGHARNEIN, H., SAVARI, A., SAFAHIEH, A.R. AND GHASEMI, A., 2012. Genetic structure four finger threadfin (Eleutheronema tetradactylum) population in the Persian Gulf and the Oman sea using RAPD marker. World Journal of Fish and Marine Sciences, 4(1), 78-82.

KOCHER, T.D., CONROY, J.A., MCKAYE, K.R. AND STAUFFER, J.R., 1993. Similar morphologies of cichlid fish in Lakes Tanganyika and evolution, edited by KEENLEYSIDE, M. H. A. Chapman and Hall, Malawi are due to convergence. Mol Phylogenet Evol, 2, 158–165.

395

KUMLA, S., DOOLGINDACHBAPORN, S., SUDMOON, R. AND SATTAYASAI, N., 2012. Genetic variation, population structure and identification of yellow catfish, Mystus nemurus (C and V) in Thailand using RAPD, ISSR and SCAR marker. Mol Biol Rep, 39, 5201–5210.

LAKRA, W.S., GOSWAMI, M. AND GOPALAKRISHNAN, A., 2009. Molecular identification and phylogenetic relationships of seven Indian Sciaenids (Pisces: Perciformes, Sciaenidae) based on 16S rRNA and Cytochrome c oxidase subunit I mitochondrial genes. Mol Biol Rep, 36, 831–839.

LANGERHANS, B. R., LAYDON, C. A., LANGERHANS, A. K., AND DEWITT, T. J., 2003. Habitat associated morphological divergence in two Neotropical fish species. Biol J of the Linn Soci London, 80, 689-698.

LASHARI, P. K., NAREJO, N. T., MASTOI, A. M. AND MAHAR M. A., 2004. Some morphometric characters and their relationship in carp, Cirrhinus reba (Hamilton) from fishpond district Jacobabad, Sindh. Proc Pakistan Congr Zool, (24), 179-184.

LEE, P.G. AND NG, P.K.L., 1994. The systematics and ecology of snakeheads (Pisces: Channidae) in Peninsular Malaysia and Singapore. Hydrobiologia, 285, 59-74.

LEHMANN, D., HETTWER, H. AND TARASCHEWSKI, H., 2000. RAPD-PCR investigations of systematic relationships among four species of eels (Teleostei: Anguillidae), particularly Anguilla anguilla and A. rostrata. Mar Biol, 137, 195-204.

LEUZZI, M.S.P., ALMEIDA, F.S. AND ORSI, M.L., 2004. Analysis by RAPD of the genetic structure of Astyanax altiparanae (Pisces: Characiformes) in reservoirs of the Paranapanema river, Brazil. Genet Mol Biol, 27, 335-362. LI, D., AND HU, X., 2009. Fish and its multiple human health effects in times of threat to sustainability and affordability: Are there alternatives? Asia Pac J Clin Nutr, 18(4), 553-563.

396

LING, S.W., 1977. Aquaculture in Southeast Asia a historical overview. Seattle University of Washington press. 108

LIU, Y.G., YU, Z.G., BAO, B.L., SAN, X.Q., SHI, Q.L. AND LIU, L.X., 2009. Population genetics studies of half-smooth tongue sole Cynog- lossus semilaevis using ISSR markers. Biochem Syst Ecol, 36, 821–827.

LIU, Z., JARRET, R.L., DUNCAN, R.R., AND KRESOVICH, S., 1994. Genetic relationships and variation among ecotypes of seashore Paspalum (Paspalum vaginatum) determined by random polymorphic DNA markers. Genome, 37, 1011–1017

LOPERA-BARRERO, N.M., POVH, J.A., RIBEIRO, R.P., GOMES, P.C., JACOMETO, C.B. AND LOPES T.D.S., 2008. Comparison of DNA extraction protocols of fish fin and larvae samples: modified salt (NaCl) extraction. Cien Inv Agr, 35(1), 65-74.

MABUCHI, K., MIYA, M., SATOH, T.P., WESTNEAT, M.W. AND NISHIDA, M. 2004. Gene rearrangements and evolution of tRNA pseudogenes in the mitochondrial genome of the parrotfish (Teleostei: Perciformes: Scaridae). J Mol Evol, 59, 287–297.

MANDAL, A., LAL, K.K., MOHINDRA, V., SINGH, R.K., PUNIA, P., CHAUHAN, U.K. AND LAKRA, W.S., 2009. Evaluation of genetic variation in the Clown Knife fish, Chitala chitala, using Allozymes, RAPD and Microsatellites. Biochem Genet, 47, 216-234.

MAT- JAIS, A.M., MCCULLOH, R. AND CROFT, K., 1994. Fatty acid and amino acid composition in haruan as a potential role in wound healing. Gen Pharmacol, 25, 947-950.

MATOSO, D.A., SILVA, M. CORTINHAS, M.C.S., CESTARI, M.M., ALMEIDA, M.C., VICARI, M.R. AND ARTONI, R.F., 2011. Two genetic stocks of Steindachneridion melanodermatum living in sympatry in nature and genetic variability

397

of wild parents and F1 generation. Genet Mol Res,

10(4), 2606-2612.

MICHELLE, N.Y.T., SHANTI, G. AND LOQMAN, M.Y., 2004. Effect of orally administered Channa striatus extract against experimentally induced osteoarthritis in rabbits. Int J Appl Res Vet Med, 2,171-175.

MIRZA, .M.R., 1982. A contribution to the Fishes of Lahore. Polymer Publications Urdu Bazaar, Lahore, 48.

MIRZA, M. R., 2003. Checklist of freshwater fishes of Paksitan. Pakistan J. Zool. Suppl. Series No. 3, 1-30.

MIRZA, M.R., AND BHATTI, M.N., 1993. Fish and fisheries of Pakistan, Part I. Lahore, Ferozsons.

MIRZA, M.R. AND BHATTI, M.N., 1995. Fish and fisheries of Pakistan, Part II. Lahore, Ferozsons.

MOHANTY, B.P., PARIA, P., DAS, D., GANGULY, S., MITRA, P., VERMA, A., SAHOO, S., MAHANTY, A., AFTABUDDIN, Md., BEHERA, B.K., SANKAR, T.V. AND SHARMA, A.P., 2012. Nutrient Profile of Giant River- Catfish Sperata seenghala (Sykes) Nat Acad Sci Lett (India), 35 (3), 155-161.

MOHINDRA, V., KHARE, P., LAL, K.K., PEYUSH, P., SINGH, K.S., BARMAN, A.S. AND LAKRA, W.S., 2007. Molecular discrimination of five Mahseer species from Indian peninsula using RAPD analysis. Acta Zool Sin, 53(4), 725–732.

MOHSIN, A.K.M. AND AMBAK, M.A., 1983. Freshwater Fishes of Peninsular Malaysia. Penerbitan Universiti Pertanian Malaysia; Kuala Lumpur, 284.

MUNEER, P.M.A, GOPALAKRISHNAN, A., MUSAMMILU, K.K., BASHEER, V.S. MOHINDRA, V., LAL, K.K., PONNIAH, A.G. AND PADMAKUMAR, K.G., 2012. Comparative assessment of genetic variability in the populations of

398

endemic and endangered Yellow Catfish, Horabagrus brachysoma (Teleostei: Horabagridae), based on Allozyme, RAPD, and Microsatellite markers. Biochem Genet, 50(3-4), 192-212.

MUNEER, P.M.A, GOPALAKRISHNAN, A., MUSAMMILU, K.K., MOHINDRA, V., LAL, K.K., BASHEER, V.S. AND LAKRA, W.S., 2009. Genetic variation and population structure of endemic yellow catfish, Horabagrus brachysoma (Bagridae) among three populations of Western Ghat region using RAPD and microsatellite markers. Mol Biol Rep, 36, 1779–1791.

MUNEER, P.M.A, GOPALAKRISHNAN, A., SHIVANANDAN, R., BASHEER, V.S. AND PONNIAH, A.G., 2011a. Genetic variation and phylogenetic relationship between two species of yellow catfish, Horabagrus brachysoma and H. nigricollaris (Teleostei: Horabagridae) based on RAPD and microsatellite markers. Mol Biol Rep, 38, 2225-2232.

MUNEER, P.M.A, SIVANANDAN, R., GOPALAKRISHNAN, A., BASHEER, V.S., MUSAMMILU, K.K. AND PONNIAH, A.G. 2011b. Development and characterization of RAPD and microsatellite markers for genetic variation analysis in the critically endangered yellow catfish Horabagrus nigricollaris (Teleostei: Horabagridae). Biochem Genet, 49, 83– 95.

MUNEER, P.M.A., GOPALAKRISHNAA, A., BASHEER, V.S. AND LAKRA, W.S., 2008. Identification of random amplified polymorphic DNA (RAPD) markers in endemic yellow catfish, Horabagrus brachysoma (Gunther 1864). Asian Fish Sci, 21(3), 293–304.

MUNRO, I.S.R., 1955. The marine and freshwater fishes of Ceylon. Dept Ext Affairs, Canbera, 348. 56 pls.

MURTA, A.G., 2000. Morphological variation of horse mackerel (Trachurus trachurus) in the Iberian and North Africa Atlantic implications for stock identification. ICES J Mar Sci, 57, 1240–8.

399

MUSIKASINTHORN, P. AND TAKI, Y., 2001. Channa siamensis (Gu¨nther, 1861), a junior synonym of Channa lucius (Cuvier n Cuvier and Valenciennes, 1831). Ichthyol Res, 48, 319–324.

MUSIKASINTHORN, P., 1998. Channa panaw, a new channid fish from the Irrawaddy and Sittang River basins, Myanmar. Ichthyol Res, 45, 355–362.

MUSIKASINTHORN, P., 2000. Channa aurantimaculata, a new channid fish from Assam (Brahmaputra River basin), India, with designation of a neotype for C. amphibeus (McClelland, 1845). Ichthyol Res, 47 27–37.

MUSIKASINTHORN, P., 2003. Channoidae (Snakeheads) In HUTCHINS, M., THONEY, A., LOISELLE, P.V. AND SCHLAGER N (eds) Grzimek’s animal life encyclopedia, 2nd edn. vols 4, 5. Fishes I–II. Gale Group, Farmington Hills, 437–447.

NAGARAJAN, M., HANIFFA, M.A., GOPALAKRISHNAN, A., BASHEER, V.S. AND MUNEER, A., 2006. Genetic variability of Channa punctatus populations using randomly polymorphic DNA. Aquacult Res, 37, 1151–1155.

NAISH, K.A. AND SKIBINSKI, D.O.F., 1998. Tetra nucleotide microsatellite loci for Indian major carp. J Fish Biol, 53, 886–889.

NAKAMURA, T., 2003. Meristic and morphometric variations in fluvial Japanese charr between river systems and among tributaries of a river system. Env Biol Fish, 66, 133–41.

NAREJO, N.T., LASHARI, P.K. AND JAFRI, S.I.H., 2008. Morphometric and meristic differences between two types of Palla, Tenualosa ilisha (Hamilton) from River Indus Pakistan. Pak J Zool, 40 (1), 31-35.

NEI, M., 1978. Estimation of average heterozygosity and genetic distance from a small number of individuals. Genetics 89, 583-590. 400

NELSON, J.S., 1984. Fishes of the World. New York John Wiley and Sons. 523.

NG, H.H. 2010. Wallago attu. In: IUCN 2012. IUCN Red List of Threatened Species. Version 2012.2. . Downloaded on 3 November 2012.

NG, P.K.L. AND LIM, K.K.P., 1990. Snakeheads (Pisces Channidae) natural history, biology and economic importance. In MING, C.L.; NG P.K.L. (eds) Essays in zoology papers commemorating the 40th anniversary of the Department of Zoology. National University of Singapore, Singapore, 127–152.

NGUYEN, T.T.T., 2008. Population structure in the highly fragmented range of Tor douronensis (Cyprinidae) in Sarawak, Malaysia revealed by microsatellite DNA markers. Fresh water Biol, 53, 924–934.

OKAZAKI, T., JEON, S., WATANABE, M. AND KITAGAVA, T., 1999. Genetic Relationships of Japanese and Korean Bagrid Catfishes from Mitochondrial DNA Analysis. Zool Sci, 16, 363–373.

OKUMUS, I. AND CIFTCI, Y., 2003. Fish population genetics and molecular markers II- molecular markers and their applications in fisheries and aquaculture. Turk J Fish Aquat Sci, 3, 51–79.

PADHI, B.K. AND MANDAL, R.K., 1995a. Genetic resource mapping and fisheries management. Curr Sci, 68(5), 490– 493.

PADHI, B.K. AND MANDAL, R.K., 1995b. Restriction site polymorphism in mitochondrial DNA of Indian major carps. Curr Sci, 68(9), 965–967.

PERVAIZ, K., IQBAL, Z., MIRZA, M.R., JAVED, M.N. AND NAEEM, M., 2012. Meristic and morphometric studies on Indus mahseer Tor macrolepis (Teleostei: Cyprinidae) from district Attock, Pakistan. Int J Agric Biol, 14, 169–175.

401

PETHIYAGODA, R., 1991. Freshwater fishes of Sri Lanka. Wildlife Heritage Trust of Sri Lanka, Colombo. 362.

PHALE, S.R., CHAUHAN, S., BHUTE, Y.V. AND BAILE, V.V., 2009. Detection of genetic variation in the wild populations of Indian major carps using random amplified polymorphic DNA fingerprinting. J Fish Aquat Sci, 4(1), 63–70.

POGSON, G.H., MESA, K.A. AND BOUTILLIER, R.G., 1995. Genetic population structure and gene flow in the Atlantic Cod Gadus morhua a comparison of allozyme and nuclear RFLP loci. Genetics, 139, 375–385.

POULET, N., BERRABI, P., CRIVELLI, A.J., LEK, S. AND ARGILLIER, C., 2004. Genetic and morphometric variations in the pikeperch (Sander lucioperca L.) of a fragmented data. Arch Hydrobiol, 159 (4), 531–54.

PRENTIS, P.J., AND MATHER, P.B., 2008. Fine-scale patterns of genetic variation indicate non-equilibrium gene frequency divergence in the stream lily, Helmholtzia glaberrima. Freshwater Biol, 53(5), 973-980.

PUNIA, P., GUPTA, H.S., SINGH, R.K., MOHINDRA, V., LAL, K.K., RANJANA, CHAUHAN, V.S AND LAKRA, W.S., 2006. Polymorphic microsatellite markers isolated from partially enriched genomic library of Chitala chitala. Mole Ecol, Notes, 6, 1263–1265.

QURESHI, M. R., 1965. Common freshwater fishes of Pakistan. Agri. Res. Council. Karachi, 1-61.

RAHMAN, A.K.A. 2005. Freshwater Fishes of Bangladesh (Second edition). The Zoological Society of Bangladesh, Department of Zoology, University of Dhaka, Dhaka-1000. 394 pp.

RAMELLA, M.S., KROTH, M.A., MEURER, S., NUÑER, A.P.O., FILHO, E.Z. AND ARISI, A.C.M., 2006. Genetic variability in four fish species (Pimelodus maculatus, Prochilodus lineatus, Salminus brasiliensis and Steindachneridion scripta) from Uruguay river basin. Braz Arch Biol Tech, 49, 589-598.

402

RAO, C.B. AND MAJUMDAR, K.C., 1998. Multivariate map representation of phylogenetic relationships application to tilapiine fish. J of Fish Biol, 52, 1199–1217.

RASHID, J., TAMANNA, F.M., HOSSAIN, M.A.R. AND ALAM, S.Md., 2012. Genetic variation in endangered butter catfish, Ompok bimaculatus (Bloch) populations revealed by random amplified polymorphic DNA (RAPD) fingerprinting. Int J Biosci, 2(9), 85-93.

RASOOL, F., QURESHI, N.A., PARVEEN, S., SIDDIQUE, M, ASHRAF, M., HAMEED, A., KHAN, N. AND IQBAL, K.J., 2012a. Morphometric Parameters Assisted Study of the Hatchery Raised and Natural Populations of Labeo rohita. Pak J of Zool, 45(4), 903-907.

RASOOL, F., QURESHI, N.A., PARVEEN, S., SIDDIQUE, M, ASHRAF, M., HAMEED, A. AND IQBAL, K.J, 2012b. Principle Component Analysis for the Variation among the Populations of Labeo rohita. Pak J Ani Plant Sci, 23(2), 478-492.

RASOOL, F., QURESHI, N.A., SIDDIQUE, M, ASHRAF, M., Ali, A., HAMEED, A. AND PARVEEN, S., 2012c. Clustering Analysis for Intraspecific Variation Studies amongst the Populations of Cirrhinus mrigala. Pak J Ani And Plant Sci, 23(5),1327-1332.

ROSHADA, H., 1994. The effect of mixed feeding schedules of varying dietary protein content on the growth performance of Channa striata fry. Asian Fish Sci., 7, 149-155.

SAIKIA, A.K., 2012. Morphometric and Biometric Index Study of Channa punctatus (Bloch) from Paddy Field of Sivsagar District, Assam. J Biol Chem Research, 29(1), 37-43.

403

SAINI, A., DUA, A. AND MOHINDRA, V., 2010. Genetic Variability Analysis of Giant River Cattish (Sperata seenghala) Populations from Indus River System by RAPD_PCR. Russ J Genet, 46(8), 982–987.

SAINI, A., DUA, A., MOHINDRA, V. AND LAKRA, W.S., 2011. Molecular discrimination of six species of Bagrid catfishes from Indus river system using randomly amplified polymorphic DNA markers. Mol Biol Rep, 38, 2961-2965.

SAINI, A., DUAA, A. AND MOHINDRAB, V., 2008. Comparative morphometrics of two populations of giant river catfish (Mystus seenghala) from the Indus river system. Integrative Zoology, 3, 219-226.

SALAM, A.A., CHOWDHRY, A., ANSARI, A. AND NADEEM, S., 1997. Studies on the effect of seasonal variations of physico-chemical parameters of Indus River. Sindh Uni Res J (Sci. Ser.), 29 (1), 41-50.

SAMANTARAY, K. AND MOHANTY, S.S., 1997. Interactions of dietary levels of protein and energy on fingerling snakehead, Channa striata. Aquaculture, 156, 241- 49.

SANDOVAL-CASTELLANOS, E., URIBE-ALCOCER, M. AND DIAZ-JAIMES, P., 2007. Population genetic structure of jumbo squid (Dosidicus gigas) evaluated by RAPD analysis. Fish Res, 83, 113–118.

SCHIERWATER, B., ENDER, A., SCHWENK, K., SPAAK, P., STREIT, B., 1994. In: Molecular ecology and evolution approaches and applications, SCHIERWATER, B., STREIT, B., WAGNER, G.P. AND DESALLE, R., eds., Basel Birkhouser. 495– 508.

SCHLUTER, D., AND MCPHAIL, J. D., 1992. Ecological character displacement and speciation in sticklebacks. American Naturalist, 140, 85-108.

SHINDE, S.E., PATHAN, K.S., RAUT, T.S., BHANDARE, R.Y. AND SONAWANE, D.I., 2009. Fish diversity of Pravara River at Pravara Sangam District Ahmed nagar, (M. S.) India. World J Zool, 4(3), 176-179.

404

SIN, F.Y.T., 1997. Transgenic fish. Rev Fish Biol Fish, 7, 417-441.

SIRAJ, S.S., ESA, Y.B., KEONG, B.P. AND DAUD, S.K., 2007. Genetic characterization of the two color-type of kelah. Malays Appl Biol, 36 (1), 23–29.

SNORRASON, S.S., SKULASON, S., JONSSON, B., MALMQUIST, H.J., JONASSON, P.M., SANDLUND, O.T., AND LINDEM, J., 1994. Trophic specialization in Arctic char Salvelinus alpinus (Pisces, Slmonidae): morphological divergence and ontogenetic niche shifts. Biological Journal of the Linnean Society of London, 52, 1-18.

STERGIOU, K.I. AND VASILIKI, S., 2003. Length–girth relationships for several marine fishes. Fish Res, 60, 161–168.

SULTANA, S., AKTER, S., HOSSAIN, M., AND ALAM, M., 2010. DNA fingerprinting of the Asian stinging catfish (Heteropneustes fossilis, Bloch) by Random Amplified Polymorphic DNA markers. Interl J Biotechnol, 2(2), 5-10.

SUN, Y., LIU, S., ZHAO, G., HE, S., WU, Q., TANIGUCHI, N. AND YU, Q., 2004. Genetic structure of Chinese sucker population Myxocyprinus asiaticus in the Yangtze River based on mitochondrial DNA marker. Fish Sci, 70, 412- 420.

SYKES, W.H., 1839 On the fishes of the Deccan. Proceedings of the General Meetings for Scientific Business of the Zoological Society of London, (pt 6) 157-165.

TALWAR, P.K. AND JHINGRAN, A.G., 1992. Inland fishes of India and adjacent countries. A. A. Alkema/Rotterdam. 1- 1062.

TAMANNA, F.M., RASHID, J. AND ALAM, S. Md., 2012. High levels of genetic variation revealed in wild populations of the stripped dwarf catfish Mystus vittatus (Bloch) (Bagridae: iluriformes) in Bangladesh by random amplified polymorphic DNA techniques. Inter J Adv Biol Res, 2(2), 322-327.

405

TEUGELS, G.G., 1992. Channidae. In: LEVEQUE, C., PAUGY, D. AND TEUGELS, G.G. (eds) The fresh and brackish water fishes of west Africa, vol 2. Coll. Faune tropicale XXVIII. ORSTOM - MRAC, Paris, 655–658.

TIANPEI, M., 1991. Anatomy and systematics of Bagridae (Teleostei) and Siluroid phylogeny. Koeltz Scientific Books, Koenigstein, 124–137.

TINTI, F., DI NUNNO C, GUARNIERO, I., TALENTI, M., TOMMASINI, S., FABBRI, E. AND PICCINETTI, C., 2002. Mitochondrial DNA sequence variation suggests the lack of genetic heterogeneity in the Adriatic and Ionian stocks of Sardina pilchardus. Mar Biotechnol, 4, 163–172.

TRIPATHI, S.D., 1996. Present status of breeding and culture of catfishes in south Asia. Aquatic Living Res, 9, 219–228.

TURAN, C., ERGUDEN, D., TURAN, F. AND GURLEK, M., 2004. Genetic and morphologic structure of Liza abu (Heckel, 1843) population from the rivers Orontes, Euphrates and Tigris. Turkish J of Vet and Ani Sci, 28, 729–34.

TURAN, C., YALCIN, S., TURAN, F., OKUR, E. AND AKYURT, I., 2005. Morphometric comparisons of African catfish, Clarias gariepinus, populations in Turkey. Folia Zoologica, 54 (1-2), 165–72.

UROOJ, N., SAHATO, G.A., LASHARI, K.H., KORAI, A.L., PALH, Z.A. AND NAQVI, H.S., 2011. Ichthyodiversity of River Indus, at Jamshoro District, Sindh. Sindh Univ Res Jour (Sci Ser), 43 (1), 13-18.

VERMA, J., AND SERAJUDDIN, M., 2012. Phylogenetic relationship between four species using divergent domain d8 in family: siluridae. Int J life sci and pha res, 2, 240-244.

VIGNAL, A., MILAN, D., SANCRISTOBAL, M., AND EGGEN, A., 2002. A review on SNP and other types of molecular markers and their use in animal genetics. Genetics, Selection, Evolution, 34(3), 275-306.

406

VISHWANATH, W. AND GEETHAKUMARI, K.H., 2009. Diagnosis and interrelationships of fishes of the genus Channa and Scopoli (Teleostei: Channidae) of northern. Ind J Threatened Taxa, 1(2), 97–105.

VRIJENHOEK, R.C., 1998. Conservation genetics of freshwater fish. J Fish Biol, 53 (Suppl A), 394–412.

WEE, K.L., AND TACON, A.G.J., 1982. A preliminary study on the dietary protein requirements of juvenile snakeheads. Bull Jpn Soc Sci Fish, 48(10), 1463-1468.

WELCOME, R. L., 1985. River fisheries. FAO fisheries technical paper No. 262. Food and Agriculture Organization of United Nations, Rome.

WILLIAMS, J.G., KUBELIK, A.R., LIVAK, K.J., RAFALSKI, J.A. AND TINGEY S.V., 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucl Acids Res, 18, 653–6535.

WOLSTENHOLME, D.R., 1992. Animal mitochondrial DNA: structure and evolution. Int Rev Cytol, 141, 173–216. YADEV, B., 2006. Indian Fisheries. In: Fish and fisheries, II edn. Daya Publishing House, Delhi, India, 183–190.

YOON, J.M. AND KIM, G.W., 2001. Randomly amplified polymorphic DNA–polymerase chain reaction analysis of two different populations of cultured Korean catfish Silurus asotus. J of Biosci, 26 (5), 641–647.

ZAKARIA, Z.A., MAT JAIS, A.M., GOHM, Y.M., SULAIMAN, M.R. AND SOMCHIT, M.N., 2007. Amino acid and fatty acid composition of an aqueous extract of Channa striatus (Bloch) that exhibits antinociceptive activity. Clin Exp Pharmacol Physiol, 34, 198-204.

ZHANG, C-G., MUSIKASINTHORN, P. AND WATANABE, K., 2002. Channa nox, a new channid fish lacking a pelvic fin from Guangxi, China. Ichthyol Res, 49, 140–146.

407

ZHANG, S. AND REDDY, P.V.G.K., 1991. On the comparative karyomorphology of three Indian major carps, Catla catla (Hamilton), Labeo rohita (Hamilton) and Cirrhinus mrigala (Hamilton). Aquaculture, 97, 7 -12.

ZHU, X.P., ZHAO, J., CHEN, K.C., LIU, Y.H., CHEN, Y.L., ZHENG, G.M. AND ZHANG, D.D., 2009. Genetic analysis of ratmouth barbell (Ptychidio jordani Myers) from different geographic sub-populations in the Pearl River Basin using microsatellite markers. J Appl Ichthyol, 25, 148–152.

ZURAINI, A, SOMCHIT, M.N, SOLIHAH, M.H, GOH, Y.M, ARIFAH, A.K, ZAKARIA, M.S, SOMCHIT, N., RAJION, M.A, ZAKARIA, Z.A. AND MAT JAIS, A.M., 2006. Fatty acid and amino acid composition of three local Malaysian Channa spp. fish. Food Chem, 97, 674-678.

408

Appendices APPENDICES

Appendix 1: Morphometric Parameters of Channa punctatus Sites Wt TL HL St DF CF PF

Blk 150.0 22.9 4.3 12.5 1.3 2.5 2.5

Blk 140.0 17.8 3.4 9.9 1.3 2.0 2.1

Blk 150.0 17.1 3.3 9.5 1.2 1.9 1.9

Blk 120.0 18.4 3.6 10.2 1.3 2.1 2.1

Blk 145.0 17.3 3.4 9.6 1.2 1.9 1.9

Blk 160.0 20.3 3.9 11.3 1.5 2.3 2.3

Blk 135.0 17.8 3.4 9.9 1.3 2.0 1.8

Blk 115.0 15.2 3.0 8.5 1.1 1.7 1.6

Blk 95.0 22.2 4.3 12.4 1.6 2.5 2.5

Blk 145.0 21.0 4.1 11.7 1.5 2.3 2.3

Average 135.5 19.0 3.7 10.6 1.3 2.1 2.1 cha 180.0 15.9 3.1 8.8 1.1 1.8 1.7 cha 140.0 15.2 3.0 8.5 1.1 1.7 1.7

409

cha 135.0 17.1 3.3 9.5 1.2 1.9 1.8 cha 145.0 12.7 2.5 7.1 0.9 1.4 1.4 cha 136.0 15.9 3.1 8.8 1.1 1.8 1.8 cha 160.0 12.1 2.3 6.7 0.9 1.4 1.3 cha 140.0 13.7 2.7 7.6 1.0 1.5 1.5 cha 120.0 16.5 3.2 9.2 1.2 1.8 1.8 cha 175.0 17.8 3.4 9.9 1.3 2.0 2.0 cha 165.0 16.5 3.2 9.2 1.2 1.8 1.6

Average 149.6 15.3 3.0 8.5 1.1 1.7 1.7

Qbd 125.0 19.1 3.7 10.6 1.4 2.1 2.1

Qbd 120.0 15.9 3.1 8.8 1.1 1.8 1.8

Qbd 135.0 12.7 2.5 7.1 0.9 1.4 1.3

Qbd 100.0 15.2 3.0 8.5 1.1 1.7 1.7

Qbd 125.0 17.8 3.4 9.9 1.3 2.0 1.9

Qbd 95.0 17.1 3.3 9.5 1.2 1.9 1.9

Qbd 108.0 19.7 3.8 10.9 1.4 2.2 2.1

Qbd 130.0 17.1 3.3 9.5 1.2 1.9 1.9

Qbd 140.0 15.2 3.0 8.5 1.1 1.7 1.7

410

Qbd 130.0 16.5 3.2 9.2 1.2 1.8 1.8

Average 120.8 16.6 3.2 9.3 1.2 1.9 1.8

Trm 170.0 19.1 3.7 10.6 1.4 2.1 2.2

Trm 168.0 17.8 3.4 9.9 1.3 2.0 2.0

Trm 160.0 19.1 3.7 10.6 1.4 2.1 2.1

Trm 180.0 15.2 3.0 8.5 1.1 1.7 1.7

Trm 140.0 18.4 3.6 10.2 1.3 2.1 2.2

Trm 135.0 20.3 3.9 11.3 1.5 2.3 2.3

Trm 140.0 17.1 3.3 9.5 1.2 1.9 1.9

Trm 100.0 14.6 2.8 8.1 1.1 1.6 1.5

Trm 90.0 12.1 2.3 6.7 0.9 1.4 1.4

Trm 110.0 18.4 3.6 10.2 1.3 2.1 2.1

Average 139.3 17.2 3.3 9.6 1.2 1.9 1.9

Tsa 150.0 21.6 4.2 12.0 1.6 2.4 2.5

Tsa 125.0 21.3 4.1 11.9 1.5 2.4 2.4

Tsa 100.0 20.3 3.9 11.3 1.5 2.3 2.3

Tsa 120.0 22.9 4.4 12.7 1.6 2.6 2.5

Tsa 140.0 17.8 3.4 9.9 1.3 2.0 2.0

411

Tsa 135.0 17.1 3.3 9.5 1.2 1.9 1.9

Tsa 155.0 17.8 3.4 9.9 1.3 2.0 2.0

Tsa 135.0 12.7 2.5 7.1 0.9 1.4 1.4

Tsa 120.0 11.4 2.2 6.4 0.8 1.3 1.4

Tsa 130.0 14.0 2.7 7.8 1.0 1.6 1.6

Average 131.0 17.7 3.4 9.8 1.3 2.0 2.0

Cha = Chashma Barrage, Qbd = Qadirabad Barrage, Blk = Baloki Barrage, Trm = Trimu Barrage, Tsa = Taunsa Barrage

Wt = Wet Body Weight, TL = Total Length, HL = Head Length, St = Stoutness, DF = Dorsal Fin, CF = Caudal Fin and PF = Pectoral Fin Appendix 2: Morphometric Parameters of Channa marulius

Sites Wt TL HL St DF CF AF PF

Blk 2550.0 60.9 14.2 27.9 3.9 10.4 3.9 9.1

Blk 1220.0 29.1 6.8 13.3 1.9 5.0 1.9 4.3

Blk 2876.0 68.7 16.1 31.4 4.4 11.7 4.4 10.2

Blk 2765.0 66.0 15.5 30.2 4.2 11.2 4.2 9.8

Blk 2534.0 60.5 14.2 27.7 3.9 10.3 3.9 9.0

Blk 2142.0 51.1 12.0 23.4 3.3 8.7 3.3 7.6

Blk 2378.0 56.8 13.3 26.0 3.6 9.7 3.6 8.5

412

Blk 1345.0 32.1 7.5 14.7 2.0 5.5 2.0 4.8

Blk 1452.0 34.7 8.1 15.9 2.2 5.9 2.2 5.2

Blk 1000.0 23.9 5.6 10.9 1.5 4.1 1.5 3.6

Average 2026.2 48.4 11.3 22.1 3.1 8.2 3.1 7.2 cha 3000.0 71.6 16.8 32.8 4.6 12.2 4.6 10.7 cha 2567.0 61.3 14.3 28.0 3.9 10.4 3.9 9.1 cha 2678.0 63.9 15.0 29.2 4.1 10.9 4.1 9.5 cha 2970.0 70.9 16.6 32.4 4.5 12.1 4.5 10.6 cha 1235.0 29.5 6.9 13.5 1.9 5.0 1.9 4.4 cha 2907.0 69.4 16.2 31.8 4.4 11.8 4.4 10.3 cha 1546.0 36.9 8.6 16.9 2.4 6.3 2.4 5.5 cha 2390.0 57.1 13.4 26.1 3.6 9.7 3.6 8.5 cha 1230.0 29.4 6.9 13.4 1.9 5.0 1.9 4.4 cha 1280.0 30.6 7.2 14.0 2.0 5.2 2.0 4.6

Average 2180.3 52.1 12.2 23.8 3.3 8.9 3.3 7.8

Qbd 1435.0 34.3 8.0 15.7 2.2 5.8 2.2 5.1

Qbd 1620.0 38.7 9.1 17.7 2.5 6.6 2.5 5.8

Qbd 2270.0 54.2 12.7 24.8 3.5 9.2 3.5 8.1

413

Qbd 2285.0 54.6 12.8 25.0 3.5 9.3 3.5 8.1

Qbd 2890.0 69.0 16.1 31.6 4.4 11.7 4.4 10.3

Qbd 1500.0 35.8 8.4 16.4 2.3 6.1 2.3 5.3

Qbd 1765.0 42.1 9.9 19.3 2.7 7.2 2.7 6.3

Qbd 1890.0 45.1 10.6 20.6 2.9 7.7 2.9 6.7

Qbd 1657.0 39.6 9.3 18.1 2.5 6.7 2.5 5.9

Qbd 1505.0 35.9 8.4 16.4 2.3 6.1 2.3 5.4

Average 1881.7 44.9 10.5 20.6 2.9 7.6 2.9 6.7

Trm 2978.0 71.1 16.6 32.5 4.5 12.1 4.5 10.6

Trm 2660.0 63.5 14.9 29.1 4.1 10.8 4.1 9.5

Trm 2646.0 63.2 14.8 28.9 4.0 10.8 4.0 9.4

Trm 1324.0 31.6 7.4 14.5 2.0 5.4 2.0 4.7

Trm 1445.0 34.5 8.1 15.8 2.2 5.9 2.2 5.1

Trm 1334.0 31.9 7.5 14.6 2.0 5.4 2.0 4.7

Trm 1768.0 42.2 9.9 19.3 2.7 7.2 2.7 6.3

Trm 1290.0 30.8 7.2 14.1 2.0 5.2 2.0 4.6

Trm 1420.0 33.9 7.9 15.5 2.2 5.8 2.2 5.0

Trm 1657.0 39.6 9.3 18.1 2.5 6.7 2.5 5.9

414

Average 1852.2 44.2 10.4 20.2 2.8 7.5 2.8 6.6

Tsa 1875.0 44.8 10.5 20.5 2.9 7.6 2.9 6.7

Tsa 1940.0 46.3 10.8 21.2 3.0 7.9 3.0 6.9

Tsa 2870.0 68.5 16.0 31.3 4.4 11.7 4.4 10.2

Tsa 2355.0 56.2 13.2 25.7 3.6 9.6 3.6 8.4

Tsa 1687.0 40.3 9.4 18.4 2.6 6.9 2.6 6.0

Tsa 1789.0 42.7 10.0 19.5 2.7 7.3 2.7 6.4

Tsa 1350.0 32.2 7.5 14.7 2.1 5.5 2.1 4.8

Tsa 1450.0 34.6 8.1 15.8 2.2 5.9 2.2 5.2

Tsa 1580.0 37.7 8.8 17.3 2.4 6.4 2.4 5.6

Tsa 1860.0 44.4 10.4 20.3 2.8 7.6 2.8 6.6

Average 1875.6 44.8 10.5 20.5 2.9 7.6 2.9 6.7

Cha = Chashma Barrage, Qbd = Qadirabad Barrage, Blk = Baloki Barrage, Trm = Trimu Barrage, Tsa = Taunsa BarrageWt = Wet Body Weight, FL = Fork Length, TL = Total Length, HL = Head Length, St = Stoutness,

DF = Dorsal Fin, CF = Caudal Fin, AF = Anal Fin and PF = Pectoral Fin Appendix 3: Morphometric Parameters of Rita rita

Sites Wt FL TL HL St DF CF AF AdF PF PeF

Blk 176 25.3 29.8 4.5 29.8 6.7 5.9 3.0 1.2 5.9 2.4

415

Blk 190 27.4 32.1 4.8 32.2 7.2 6.4 3.2 1.3 6.4 2.6

Blk 143 20.6 24.2 3.6 24.2 5.4 4.8 2.4 0.9 4.8 1.9

Blk 135 19.4 22.8 3.4 22.9 5.1 4.6 2.3 0.9 4.6 1.8

Blk 176 25.3 29.8 4.5 29.8 6.7 5.9 3.0 1.2 5.9 2.4

Blk 186 26.8 31.5 4.7 31.5 7.1 6.3 3.2 1.2 6.3 2.5

Blk 193 27.8 32.6 4.9 32.7 7.4 6.5 3.3 1.3 6.5 2.6

Blk 233 33.6 39.4 5.9 39.5 8.9 7.9 4.0 1.5 7.9 3.2

Blk 225 32.4 38.1 5.7 38.1 8.6 7.6 3.8 1.5 7.6 3.0

Blk 136 19.6 23.0 3.5 23.0 5.2 4.6 2.3 0.9 4.6 1.8

Average 179.3 25.8 30.3 4.6 30.4 6.8 6.1 3.1 1.2 6.1 2.4 cha 150 21.6 25.4 3.8 25.4 5.7 5.1 2.6 1.0 5.1 2.0 cha 153 22.0 25.9 3.9 25.9 5.8 5.2 2.6 1.0 5.2 2.1 cha 140 20.2 23.7 3.6 23.7 5.3 4.7 2.4 0.9 4.7 1.9 cha 102 14.7 17.3 2.6 17.3 3.9 3.4 1.7 0.7 3.4 1.4 cha 200 28.8 33.8 5.1 33.9 7.6 6.8 3.4 1.3 6.8 2.7 cha 205 29.5 34.7 5.2 34.7 7.8 6.9 3.5 1.4 6.9 2.8 cha 145 20.9 24.5 3.7 24.6 5.5 4.9 2.5 1.0 4.9 2.0 cha 140 20.2 23.7 3.6 23.7 5.3 4.7 2.4 0.9 4.7 1.9

416

cha 125 18.0 21.1 3.2 21.2 4.8 4.2 2.1 0.8 4.2 1.7 cha 132 19.0 22.3 3.4 22.4 5.0 4.5 2.2 0.9 4.5 1.8

Average 149.2 21.5 25.2 3.8 25.3 5.7 5.0 2.5 1.0 5.0 2.0

Qbd 178 25.6 30.1 4.5 30.2 6.8 6.0 3.0 1.2 6.0 2.4

Qbd 130 18.7 22.0 3.3 22.0 5.0 4.4 2.2 0.9 4.4 1.8

Qbd 132 19.0 22.3 3.4 22.4 5.0 4.5 2.2 0.9 4.5 1.8

Qbd 140 20.2 23.7 3.6 23.7 5.3 4.7 2.4 0.9 4.7 1.9

Qbd 156 22.5 26.4 4.0 26.4 5.9 5.3 2.7 1.0 5.3 2.1

Qbd 187 26.9 31.6 4.7 31.7 7.1 6.3 3.2 1.2 6.3 2.5

Qbd 189 27.2 32.0 4.8 32.0 7.2 6.4 3.2 1.2 6.4 2.6

Qbd 165 23.8 27.9 4.2 28.0 6.3 5.6 2.8 1.1 5.6 2.2

Qbd 125 18.0 21.1 3.2 21.2 4.8 4.2 2.1 0.8 4.2 1.7

Qbd 124 17.9 21.0 3.1 21.0 4.7 4.2 2.1 0.8 4.2 1.7

Average 152.6 22.0 25.8 3.9 25.9 5.8 5.2 2.6 1.0 5.2 2.1

Trm 187 26.9 31.6 4.7 31.7 7.1 6.3 3.2 1.2 6.3 2.5

Trm 109 15.7 18.4 2.8 18.5 4.2 3.7 1.9 0.7 3.7 1.5

Trm 123 17.7 20.8 3.1 20.8 4.7 4.2 2.1 0.8 4.2 1.7

Trm 154 22.2 26.1 3.9 26.1 5.9 5.2 2.6 1.0 5.2 2.1

417

Trm 167 24.1 28.3 4.2 28.3 6.4 5.6 2.8 1.1 5.6 2.3

Trm 178 25.6 30.1 4.5 30.2 6.8 6.0 3.0 1.2 6.0 2.4

Trm 124 17.9 21.0 3.1 21.0 4.7 4.2 2.1 0.8 4.2 1.7

Trm 145 20.9 24.5 3.7 24.6 5.5 4.9 2.5 1.0 4.9 2.0

Trm 176 25.3 29.8 4.5 29.8 6.7 5.9 3.0 1.2 5.9 2.4

Trm 198 28.5 33.5 5.0 33.5 7.5 6.7 3.4 1.3 6.7 2.7

Average 156.1 22.5 26.4 4.0 26.4 5.9 5.3 2.7 1.0 5.3 2.1

Tsa 132 19.0 22.3 3.4 22.4 5.0 4.5 2.2 0.9 4.5 1.8

Tsa 165 23.8 27.9 4.2 28.0 6.3 5.6 2.8 1.1 5.6 2.2

Tsa 187 26.9 31.6 4.7 31.7 7.1 6.3 3.2 1.2 6.3 2.5

Tsa 182 26.2 30.8 4.6 30.8 6.9 6.1 3.1 1.2 6.1 2.5

Tsa 193 27.8 32.6 4.9 32.7 7.4 6.5 3.3 1.3 6.5 2.6

Tsa 197 28.4 33.3 5.0 33.4 7.5 6.7 3.4 1.3 6.7 2.7

Tsa 132 19.0 22.3 3.4 22.4 5.0 4.5 2.2 0.9 4.5 1.8

Tsa 155 22.3 26.2 3.9 26.3 5.9 5.2 2.6 1.0 5.2 2.1

Tsa 145 20.9 24.5 3.7 24.6 5.5 4.9 2.5 1.0 4.9 2.0

Tsa 139 20.0 23.5 3.5 23.5 5.3 4.7 2.4 0.9 4.7 1.9

Average 162.7 23.4 27.5 4.1 27.6 6.2 5.5 2.8 1.1 5.5 2.2

Cha = Chashma Barrage, Qbd = Qadirabad Barrage, Blk = Baloki Barrage, Trm = Trimu Barrage, Tsa = Taunsa Barrage

Wt = Wet Body Weight, FL = Fork Length, TL = Total Length, HL = Head418 Length, St = Stoutness, DF = Dorsal Fin, CF = Caudal Fin, PF = Pectoral

FinandPeF = Pelvic Fin

Appendix 4: Morphometric Parameters of Sperata seenghala

Sites Wt FL TL HL St DF CF AF AdF PF PeF

Blk 1250.0 24.8 30.8 7.0 13.0 4.1 6.0 2.5 1.6 2.9 2.5

Blk 1320.0 26.2 32.5 7.4 13.7 4.4 6.4 2.7 1.7 3.0 2.7

Blk 1390.0 27.5 34.2 7.8 14.5 4.6 6.7 2.8 1.8 3.2 2.8

Blk 1700.0 33.7 41.9 9.5 17.7 5.6 8.2 3.5 2.2 3.9 3.5

Blk 1865.0 36.9 45.9 10.4 19.4 6.2 9.0 3.8 2.4 4.3 3.8

Blk 1940.0 38.4 47.8 10.8 20.2 6.4 9.4 3.9 2.5 4.4 3.9

Blk 1790.0 35.5 44.1 10.0 18.6 5.9 8.6 3.6 2.3 4.1 3.6

Blk 1260.0 25.0 31.0 7.0 13.1 4.2 6.1 2.6 1.6 2.9 2.6

Blk 1200.0 23.8 29.6 6.7 12.5 4.0 5.8 2.4 1.5 2.7 2.4

Blk 1510.0 29.9 37.2 8.4 15.7 5.0 7.3 3.1 1.9 3.5 3.1

Average 1522.5 30.2 37.5 8.5 15.9 5.1 7.3 3.1 1.9 3.5 3.1 cha 2200.0 43.6 54.2 12.3 22.9 7.3 10.6 4.5 2.8 5.0 2.4 cha 2280.0 45.2 56.2 12.7 23.7 7.6 11.0 4.6 2.9 5.2 2.6 cha 2050.0 40.6 50.5 11.5 21.3 6.8 9.9 4.2 2.6 4.7 2.7

419

cha 1800.0 35.7 44.3 10.1 18.7 6.0 8.7 3.7 2.3 4.1 3.0 cha 1750.0 34.7 43.1 9.8 18.2 5.8 8.4 3.6 2.2 4.0 3.0 cha 2430.0 48.1 59.9 13.6 25.3 8.1 11.7 4.9 3.1 5.6 2.7 cha 1400.0 27.7 34.5 7.8 14.6 4.6 6.8 2.8 1.8 3.2 3.2 cha 1620.0 32.1 39.9 9.1 16.9 5.4 7.8 3.3 2.1 3.7 3.8 cha 1730.0 34.3 42.6 9.7 18.0 5.7 8.3 3.5 2.2 4.0 4.3 cha 2165.0 42.9 53.3 12.1 22.5 7.2 10.4 4.4 2.7 4.9 3.9

Average 1942.5 38.5 47.9 10.9 20.2 6.4 9.4 3.9 2.5 4.4 3.2

Qbd 1200.0 23.8 29.6 6.7 12.5 4.0 5.8 2.4 1.5 2.7 2.4

Qbd 1270.0 25.2 31.3 7.1 13.2 4.2 6.1 2.6 1.6 2.9 2.6

Qbd 1309.0 25.9 32.3 7.3 13.6 4.3 6.3 2.7 1.7 3.0 2.7

Qbd 1480.0 29.3 36.5 8.3 15.4 4.9 7.1 3.0 1.9 3.4 3.0

Qbd 1500.0 29.7 37.0 8.4 15.6 5.0 7.2 3.0 1.9 3.4 3.0

Qbd 1350.0 26.7 33.3 7.5 14.1 4.5 6.5 2.7 1.7 3.1 2.7

Qbd 1590.0 31.5 39.2 8.9 16.6 5.3 7.7 3.2 2.0 3.6 3.2

Qbd 1850.0 36.7 45.6 10.3 19.3 6.1 8.9 3.8 2.3 4.2 3.8

Qbd 2100.0 41.6 51.7 11.7 21.9 7.0 10.1 4.3 2.7 4.8 4.3

Qbd 1940.0 38.4 47.8 10.8 20.2 6.4 9.4 3.9 2.5 4.4 3.9

420

Average 1558.9 30.9 38.4 8.7 16.2 5.2 7.5 3.2 2.0 3.6 3.2

Trm 2300.0 45.6 56.7 12.9 24.0 7.6 11.1 4.7 2.9 5.3 4.7

Trm 2345.0 46.5 57.8 13.1 24.4 7.8 11.3 4.8 3.0 5.4 4.8

Trm 2490.0 49.3 61.3 13.9 25.9 8.3 12.0 5.1 3.2 5.7 5.1

Trm 2600.0 51.5 64.1 14.5 27.1 8.6 12.5 5.3 3.3 5.9 5.3

Trm 3000.0 59.4 73.9 16.8 31.2 10.0 14.5 6.1 3.8 6.9 6.1

Trm 1680.0 33.3 41.4 9.4 17.5 5.6 8.1 3.4 2.1 3.8 3.4

Trm 1250.0 24.8 30.8 7.0 13.0 4.1 6.0 2.5 1.6 2.9 2.5

Trm 2400.0 47.5 59.1 13.4 25.0 8.0 11.6 4.9 3.0 5.5 4.9

Trm 2670.0 52.9 65.8 14.9 27.8 8.9 12.9 5.4 3.4 6.1 5.4

Trm 2200.0 43.6 54.2 12.3 22.9 7.3 10.6 4.5 2.8 5.0 4.5

Average 2293.5 45.4 56.5 12.8 23.9 7.6 11.1 4.7 2.9 5.2 4.7

Tsa 2600.0 51.5 64.1 14.5 27.1 8.6 12.5 5.3 3.3 5.9 5.3

Tsa 1500.0 29.7 37.0 8.4 15.6 5.0 7.2 3.0 1.9 3.4 3.0

Tsa 1750.0 34.7 43.1 9.8 18.2 5.8 8.4 3.6 2.2 4.0 3.6

Tsa 1930.0 38.2 47.6 10.8 20.1 6.4 9.3 3.9 2.5 4.4 3.9

Tsa 1840.0 36.5 45.3 10.3 19.2 6.1 8.9 3.7 2.3 4.2 3.7

Tsa 2070.0 41.0 51.0 11.6 21.6 6.9 10.0 4.2 2.6 4.7 4.2

421

Tsa 2300.0 45.6 56.7 12.9 24.0 7.6 11.1 4.7 2.9 5.3 4.7

Tsa 1850.0 36.7 45.6 10.3 19.3 6.1 8.9 3.8 2.3 4.2 3.8

Tsa 2400.0 47.5 59.1 13.4 25.0 8.0 11.6 4.9 3.0 5.5 4.9

Tsa 2650.0 52.5 65.3 14.8 27.6 8.8 12.8 5.4 3.4 6.1 5.4

Average 2089.0 41.4 51.5 11.7 21.8 6.9 10.1 4.2 2.7 4.8 4.2

Cha = Chashma Barrage, Qbd = Qadirabad Barrage, Blk = Baloki Barrage, Trm = Trimu Barrage, Tsa = Taunsa Barrage

Wt = Wet Body Weight, FL = Fork Length, TL = Total Length, HL = Head Length, St = Stoutness, DF = Dorsal Fin, CF = Caudal Fin, AF = Anal Fin, AdF Appendix= Adipose 5: Morphometric Fin, PF = Pectoral Parameters FinandPeF = of Pelvic Wallago Fin attu

Sites Wt FL TL HL St DF CF PF PeF

Blk 2876 66.5 72.3 16.1 26.3 8.8 6.6 8.8 4.4

Blk 1540 35.6 38.7 8.6 14.1 4.7 3.5 4.7 2.3

Blk 1345 31.1 33.8 7.5 12.3 4.1 3.1 4.1 2.0

Blk 1789 41.4 45.0 10.0 16.4 5.5 4.1 5.5 2.7

Blk 2022 46.7 50.8 11.3 18.5 6.2 4.6 6.2 3.1

Blk 1540 35.6 38.7 8.6 14.1 4.7 3.5 4.7 2.3

Blk 1450 33.5 36.5 8.1 13.3 4.4 3.3 4.4 2.2

Blk 2155 49.8 54.2 12.0 19.7 6.6 4.9 6.6 3.3

Blk 2467 57.0 62.0 13.8 22.6 7.5 5.6 7.5 3.8

422

Blk 2768 64.0 69.6 15.5 25.3 8.4 6.3 8.4 4.2

Average 1995.2 46.1 50.2 11.1 18.2 6.1 4.6 6.1 3.0 cha 1200 27.7 30.2 6.7 11.0 3.7 2.7 3.7 1.8 cha 1340 31.0 33.7 7.5 12.3 4.1 3.1 4.1 2.0 cha 1568 36.2 39.4 8.8 14.3 4.8 3.6 4.8 2.4 cha 2400 55.5 60.4 13.4 21.9 7.3 5.5 7.3 3.7 cha 2830 65.4 71.2 15.8 25.9 8.6 6.5 8.6 4.3 cha 1670 38.6 42.0 9.3 15.3 5.1 3.8 5.1 2.5 cha 1870 43.2 47.0 10.4 17.1 5.7 4.3 5.7 2.8 cha 2080 48.1 52.3 11.6 19.0 6.3 4.8 6.3 3.2 cha 2345 54.2 59.0 13.1 21.4 7.1 5.4 7.1 3.6 cha 2950 68.2 74.2 16.5 27.0 9.0 6.7 9.0 4.5

Average 2025.3 46.8 50.9 11.3 18.5 6.2 4.6 6.2 3.1

Qbd 2888 66.8 72.6 16.1 26.4 8.8 6.6 8.8 4.4

Qbd 2025 46.8 50.9 11.3 18.5 6.2 4.6 6.2 3.1

Qbd 1600 37.0 40.2 8.9 14.6 4.9 3.7 4.9 2.4

Qbd 1550 35.8 39.0 8.7 14.2 4.7 3.5 4.7 2.4

Qbd 1730 40.0 43.5 9.7 15.8 5.3 4.0 5.3 2.6

423

Qbd 1870 43.2 47.0 10.4 17.1 5.7 4.3 5.7 2.8

Qbd 1340 31.0 33.7 7.5 12.3 4.1 3.1 4.1 2.0

Qbd 1255 29.0 31.6 7.0 11.5 3.8 2.9 3.8 1.9

Qbd 2098 48.5 52.8 11.7 19.2 6.4 4.8 6.4 3.2

Qbd 2890 66.8 72.7 16.1 26.4 8.8 6.6 8.8 4.4

Average 1924.6 44.5 48.4 10.8 17.6 5.9 4.4 5.9 2.9

Trm 2567 59.3 64.5 14.3 23.5 7.8 5.9 7.8 3.9

Trm 2870 66.3 72.2 16.0 26.2 8.7 6.6 8.7 4.4

Trm 2367 54.7 59.5 13.2 21.6 7.2 5.4 7.2 3.6

Trm 2980 68.9 74.9 16.7 27.2 9.1 6.8 9.1 4.5

Trm 2976 68.8 74.8 16.6 27.2 9.1 6.8 9.1 4.5

Trm 2142 49.5 53.9 12.0 19.6 6.5 4.9 6.5 3.3

Trm 1566 36.2 39.4 8.8 14.3 4.8 3.6 4.8 2.4

Trm 1789 41.4 45.0 10.0 16.4 5.5 4.1 5.5 2.7

Trm 2060 47.6 51.8 11.5 18.8 6.3 4.7 6.3 3.1

Trm 2166 50.1 54.5 12.1 19.8 6.6 5.0 6.6 3.3

Average 2348.3 54.3 59.1 13.1 21.5 7.2 5.4 7.2 3.6

Tsa 2755 63.7 69.3 15.4 25.2 8.4 6.3 8.4 4.2

424

Tsa 1235 28.5 31.1 6.9 11.3 3.8 2.8 3.8 1.9

Tsa 1250 28.9 31.4 7.0 11.4 3.8 2.9 3.8 1.9

Tsa 1680 38.8 42.2 9.4 15.4 5.1 3.8 5.1 2.6

Tsa 2180 50.4 54.8 12.2 19.9 6.6 5.0 6.6 3.3

Tsa 1890 43.7 47.5 10.6 17.3 5.8 4.3 5.8 2.9

Tsa 1880 43.5 47.3 10.5 17.2 5.7 4.3 5.7 2.9

Tsa 1920 44.4 48.3 10.7 17.6 5.9 4.4 5.9 2.9

Tsa 1155 26.7 29.0 6.5 10.6 3.5 2.6 3.5 1.8

Tsa 1230 28.4 30.9 6.9 11.2 3.7 2.8 3.7 1.9

Average 1717.5 39.7 43.2 9.6 15.7 5.2 3.9 5.2 2.6

Cha = Chashma Barrage, Qbd = Qadirabad Barrage, Blk = Baloki Barrage, Trm = Trimu Barrage, Tsa = Taunsa Barrage

Wt = Wet Body Weight, FL = Fork Length, TL = Total Length, HL = Head Length, St = Stoutness, DF = Dorsal Fin, CF = Caudal Fin, AF = Anal Fin, AdF = Adipose Fin, PF = Pectoral FinandPeF = Pelvic Fin

425

Appendix 6: Correlations Matrix Morphometric Parameters of Channa punctatus

Wt TL HL St DF CF

TL 1.000

0.000

HL 0.991 0.988

0.000 0.000

St 0.999 0.999 0.988

0.000 0.000 0.000

DF 0.902 0.904 0.886 0.907

0.000 0.000 0.000 0.000

CF 0.974 0.974 0.964 0.976 0.904

0.000 0.000 0.000 0.000 0.000

PF 0.920 0.919 0.911 0.917 0.835 0.952

0.000 0.000 0.000 0.000 0.000 0.000

Cell Contents: Pearson correlation

P-Value

1

Appendix 7: Correlation Matrix of Morphometric Parameters of and Channa marulius

Wt FL TL HL St DF CF AF FL

1.000

0.000

TL 1.000 1.000

0.000 0.000

HL 1.000 1.000 1.000

0.000 0.000 0.000

St 1.000 1.000 1.000 1.000

0.000 0.000 0.000 0.000

DF 0.997 0.997 0.997 0.997 0.997

0.000 0.000 0.000 0.000 0.000

CF 1.000 1.000 1.000 0.999 0.999 0.996

0.000 0.000 0.000 0.000 0.000 0.000

2

AF 0.997 0.997 0.997 0.997 0.997 1.000 0.996

0.000 0.000 0.000 0.000 0.000 0.000

PF 1.000 1.000 0.999 0.999 1.000 0.997 0.999 0.997

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Cell Contents: Pearson correlation

P-Value

Appendix 8: Correlations Matrix of Morphometric Parameters of Rita rita

Wt FL TL HL St DF CF AF AdF PF

FL 1.000

0.000

TL1.000 1.000

0.000 0.000

3

HL0.995 0.995 0.995

0.000 0.000 0.000

St1.000 1.000 1.000 0.995

0.000 0.000 * 0.000

DF0.998 0.998 0.998 0.993 0.998

0.000 0.000 0.000 0.000 0.000

CF 0.997 0.996 0.996 0.992 0.996 0.995

0.000 0.000 0.000 0.000 0.000 0.000

AF0.991 0.992 0.992 0.986 0.992 0.989 0.988

0.000 0.000 0.000 0.000 0.000 0.000 0.000

AdF 0.944 0.943 0.944 0.946 0.944 0.933 0.937 0.939

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

PF0.997 0.996 0.996 0.992 0.996 0.995 1.000 0.988 0.937

0.000 0.000 0.000 0.000 0.000 0.000 * 0.000 0.000

PeF 0.983 0.983 0.983 0.979 0.983 0.984 0.975 0.973 0.897 0.975

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Cell Contents: Pearson correlation

P-Value

4

Appendix 9: Correlation Matrix of Morphometric Parameters of Sperata seenghala

Wt FL TL HL St DF CF AF AdF PF

FL 1.000

0.000

TL 1.000 1.000

0.000 0.000

HL 1.000 0.999 1.000

0.000 0.000 0.000

St1.000 1.000 1.000 0.999

0.000 0.000 0.000 0.000

DF 0.999 0.999 0.999 0.998 0.999

0.000 0.000 0.000 0.000 0.000

CF1.000 0.999 0.999 0.999 0.999 0.998

0.000 0.000 0.000 0.000 0.000 0.000

AF0.997 0.997 0.997 0.996 0.997 0.996 0.996

0.000 0.000 0.000 0.000 0.000 0.000 0.000

5

AdF 0.992 0.992 0.992 0.991 0.992 0.991 0.993 0.989

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

PF 0.997 0.997 0.997 0.996 0.996 0.996 0.997 0.994 0.990

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

PeF 0.997 0.997 0.997 0.996 0.997 0.996 0.996 1.000 0.989 0.994

0.000 0.000 0.000 0.000 0.000 0.000 0.000 * 0.000 0.000

Cell Contents: Pearson correlation

P-Value

Appendix 10: Correlations Matrix of Morphometric Parameters of Wallago attu

Wt FL TL HL St DF CF PF

FL 1.000

0.000

TL1.000 1.000

0.000 0.000

HL1.000 1.000 1.000

0.000 0.000 0.000

6

St1.000 1.000 1.000 1.000

0.000 0.000 0.000 0.000

DF0.999 0.999 0.999 0.999 0.999

0.000 0.000 0.000 0.000 0.000

CF0.999 0.999 0.999 0.999 0.999 0.998

0.000 0.000 0.000 0.000 0.000 0.000

PF0.999 0.999 0.999 0.999 0.999 1.000 0.998

0.000 0.000 0.000 0.000 0.000 * 0.000

PlF 0.996 0.996 0.996 0.996 0.997 0.994 0.994 0.994

0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000

Cell Contents: Pearson correlation

P-Value

7