FISHERY BULLETIN: VOL. 81. NO. 1 



Figure 1.— Sample locations for tile- 

 fish along the U.S. east coast and the 

 Gulf of Mexico. Submarine canyons 

 are identified in the inset. 



Electrophoresis 



Eye, liver, and muscle tissues were removed 

 from individual fish and frozen as soon as pos- 

 sible. Vertical starch gel electrophoresis was 

 used to detect protein variation. Initially only tis- 

 sues of fish from the most distant collection lo- 

 calities (Hudson Canyon and off Texas) were 

 screened for 28 enzymes to maximize the chance 

 of finding polymorphic enzymes. Of the 28 en- 

 zymes screened during the initial electrophore- 

 sis, several were scorable; however, most ap- 

 peared monomorphic (malate dehydrogenase, 

 lactate dehydrogenase, xanthine dehydrogenase, 

 creatin kinase, adenylate kinase, peptidase, alco- 

 hol dehydrogenase, malic enzyme, 6-phosphoglu- 

 conate dehydrogenase, and glyceraldehyde 3- 

 phosphate dehydrogenase) and only two [liver 

 isocitrate dehydrogenase (IDH) and liver ester- 

 ase (EST)] were polymorphic. Liver tissues from 

 all collections were then run for both IDH and 

 EST with an amine citrate buffer (pH 6.0) (Clay- 

 ton and Tretiak 1972) for 17 h at 140 V and 40°C, 

 and allelic frequencies were determined for all 

 populations. Allelic frequencies were compared 

 with their Hardy-Weinberg expectations by a 

 chi-square test (Spiess 1977). We evaluated dif- 

 ferences between sample locations, by chi-square 



contingency tests of electromorph distribution 

 between sample locations. This test does not as- 

 sume Hardy-Weinberg equilibrium and com- 

 pares n samples with k classes to determine 

 whether the individual A" classes are in the same 

 relative proportion throughout the n samples. 



Length (age)-related differences in genotype 

 distribution were tested (chi-square) on the 

 largest sample with a wide range of sizes ( n = 40, 

 west side of Hudson Canyon). Fish were divided 

 into two size classes (<550 mm fork length and 

 >550 mm) based on the approximate size at sex- 

 ual maturity. 



Morphology 



Seven meristic (number of dorsal fin spines 

 and rays, anal fin spines and rays, pectoral fin 

 rays, upper and lower gill rakers on the first 

 arch) and 21 morphometric (fork, standard, 

 total, pectoral fin, pelvic fin, upper jaw, snout, 

 adipose flap, barbel, snout to vent, snout to anal 

 origin, snout to dorsal origin, snout to incurrent 

 nostril, lengths; orbit diameter, interorbital 

 width, head width, height of first, second, and 

 third dorsal fin spines, caudal peduncle depth, 

 and suborbital depth) characters were counted or 

 measured following Hubbs and Lagler (1967), 



42 



