692 



Fishery Bulletin 91(4). 1993 



Figure 1 



Thunnus albacares. Location of sampling sites used in this analysis: (11 Manta, Ecuador (ECU); (2) the Revillagigedo Islands, Mexico 

 iMEXl; (3) Oahu, Hawaii (HAW); 1 4 1 New South Wales, Australia (AUS); (5) Manus Island, Papua New Guinea (PNG); and (6) Hatteras, 

 North Carolina (ATL). 



use of 1.5% sodium dodecyl sulfate for mitochondria] 

 lysis. Following restriction enzyme digestion and hori- 

 zontal agarose gel electrophoresis, DNA fragments were 

 transferred to nylon membranes by Southern transfer 

 (Sambrook et al., 1989) and immobilized by long-wave 

 UV irradiation. Prehybridization was conducted for two 

 hours at 42°C in 50% formamide, 5x SSC, 5x 

 Dendhardt's solution, 0.025 mM NaP0 4 , pH 6.5, and 

 100 ug/mL heat denatured calf thymus DNA. Probe 

 DNA (mtDNA purified from extra specimens from the 

 Atlantic) was nick-translated to incorporate biotin-7- 

 dATP (BRL), and separated from unincorporated nucle- 

 otides by size exclusion chromatography. One (ig of 

 probe was added to the prehybridization solution for 

 each 200 cm 2 blot and allowed to hybridize overnight 

 at 42°C. Following post-hybridization washes 

 (Sambrook et al., 1989), mtDNA fragments were visu- 

 alized with the BRL BluGene Non-Radioactive Nucleic 

 Acid Detection Kit. 



A 12-letter composite mtDNA genotype, indicating 

 the fragment pattern for each restriction enzyme, was 

 developed for each individual. Estimates of nucleon 

 diversity (h ) for each sample and for the pooled samples 

 were computed following Nei (1987). Nucleotide se- 

 quence divergences among genotypes were estimated 

 by using the site method of Nei and Li (1979). Esti- 

 mates of nucleotide sequence diversity (p) within each 

 sample, and mean nucleotide sequence divergences 

 among samples (corrected for within-sample diversi- 

 ties) were computed following Nei (1987). Nucleotide 

 sequence divergences were clustered by the unweighted 

 pair-group method with arithmetic means (UPGMA) 

 by using the average linkage algorithm of the SPSS-X 

 statistical package (Norusis, 1988). Values of G„, a 

 measure of heterogeneity between samples, were esti- 

 mated from sample genotype frequencies (Nei, 1987), 

 and values of Njn. the absolute number of migrants 

 between samples, were determined from the relation 



