McDowell and Graves; Nuclear and mitochondrial DNA markers for identification of istiophorid and xiphiid billfisfies 



539 



GTG agarose (FMC BioProducts, Rockland, ME), and vi- 

 sualized under UV light after having been stained with 

 ethidium bromide. Fragment sizes were estimated by 

 comparison with a 1-kb size standard (Life Technologies 

 Inc., Bethesda, MD) using RFLPScan Plus 3.0 (Scanalyt- 

 ics, Billeriea. MA). 



Results 



Mitochondrial marker 



Four mtDNA regions (cytochrome b. D-loop, ND4, and 

 ATPase) were included in the initial screening (Table 2). 

 The ATPase region was tested with eight potentially use- 

 ful enzjTnes based on published sequences and was found 



to have an extremely low level of interspecific variation 

 (many species exhibiting identical banding patterns). The 

 cytochrome b region was screened with four enzymes 

 based on published sequences, but because of the small 

 size of the amplification product (350bp), differences in 

 banding patterns were small and difficult to distinguish. 

 The D-loop region was screened with a total of 40 enzymes. 

 Of these, Bel I, Ahi I, Rsa I, and Hinf I were tested with 

 up to 50 individuals from each species. Banding patterns 

 that were initially thought to be diagnostic for blue marlin 

 based on Rsa I were found to occur at low frequency in 

 sailfish. This overlap combined with the large amount of 

 intraspecific variation in some species made this region 

 unsuitable for use as a forensic marker. Finally, the ND4 

 region was screened with a total of 47 restriction enzymes. 

 Of these, 17 were tested more extensively, and the combi- 



