28 



Fishery Bulletin 98(1) 



collecting the samples. Posters and information notes, 

 in Japanese and English, were distributed through- 

 out the fishery. Rewards were offered for the return 

 of tags and a substantial bonus for allowing scien- 

 tists to sample otoliths fi-om orange-tagged fish. To aid 

 in otolith recovery, kits containing large orange disks 

 were distributed to the crew of Japanese vessels; when 

 an orange-tagged fish was caught, the disks were at- 

 tached to the fish to clearly identify it in the freezers. 

 Given the high value of SBT on the Japanese 

 Sashimi market, it was essential that otoliths could 

 be sampled without affecting the external appearance 

 of the fish. Using a modification of the technique 

 described by Thorogood (1986). we removed from the 

 skull 35^4 mm cores that contained the semicircular 

 canals and sagittal otoliths with a holesaw attached 

 to a drill. The points of entry for the cores were over 

 the basioccipital plates, which are anterior to the first 

 vertebra and immediately lateral to the junction of the 

 skull and first vertebra; this area was exposed when 

 the tuna were cleaned and dressed — a process which 

 removes the gill-filaments and surrounding tissue 

 and most of the opercular flaps. By drilling through 

 each of the plates in the direction of the back of the 

 opposite eye, the cores coalesced at a point beyond the 

 sagittal otoliths and could be removed easily from the 

 skull, leaving no external mark on the fish (Fig. 1) 

 Sagittal otoliths were removed from the cores, cleaned 

 in distilled water, and dried at 28°C. 



Age estimates 



An age estimate was made from the otoliths before 

 we attempted to identify a Sr mark. Increments on 

 the whole sagittal otoliths comprised two zones: an 

 opaque (assumed to be fast growth) and a narrower 

 translucent zone (assumed to be slow growth) that ap- 

 peared dark under a dissecting microscope with re- 

 flected lighting and a black background. Following 



Thorogood's (1987) method, one of each pair of sagit- 

 tae was burned on a 400°C hot plate until it turned 

 golden brown. The color change was greater in the 

 translucent zones, making them more visible (Fig. 

 2A). Adigital image of the burnt otolith was taken (us- 

 ing the public domain "NIH Image" program" and a 

 video camera mounted on a Wild MSA dissecting mi- 

 croscope) and measurements were made of the otolith 

 length and of the distance between the primordium 

 and the inside of the translucent zones along both the 

 postrostral (PR) and transverse axes. We use termi- 

 nology that is currently widely accepted (Secor et al. 

 [1992]; KaHsh et al. [1995]; Stequert et al. [1996]). For 

 each specimen, the reader made three independent 

 age estimates by counting the number of increments 

 obvious on the distal surface of the sagitta. These were 

 made without reference to either the length of the fish 

 or the time that the fish was at liberty after tagging 

 and injection. 



Detection of strontium marks 



Scanning electron microscope (SEM) with a Robinson 

 backscatter detector In the early stages of the project 

 we used a Robinson backscatter detector coupled to a 

 Philips 515 SEM for detecting the strontium-rich band 

 in the otoliths (which we refer to as "the strontium 

 mark"). The Robinson detector visualizes differences in 

 the total atomic weight ( Z ) across a specimen . Because a 

 strontium mark within the calcium carbonate contains 

 a higher concentration of Sr, and hence a higher Z 

 than surrounding uncontaminated calcium carbonate, 

 it appears as a weak to intense bright band across 

 the growth axis of the section (Figs. 2B and 3). The 



Rasband, W. 1994. NIH Image, vers. 1.54. U.S. National Insti- 

 tutes of Health. [Available from the Internet by anonymous ftp 

 from zippy.nimh.nih.gov or as a floppy disk from NTIS, 5285 Port 

 Royal Rd., Springfield, VA 22161, part number PB93-504868.) 



