NOTE Zimmerman and Nielsen: Measurement of strontium-to-calcium ratios in otoliths of anadromous salmonids 



713 



levels; 2) we then compared the results of repeated Sr/Ca 

 measurements collected at the same spots using various 

 beam diameters, while holding accelerating voltage and 

 beam current constant to determine the effect of beam 

 damage on Sr/Ca measurements; and 3) we compared the 

 results of repeated Sr/Ca measurements collected at the 

 same spots using various beam currents, while holding ac- 

 celerating voltage and beam diameter constant. We argue 

 that increased precision of Sr measurements afforded by 

 higher beam current (and hence, higher beam power densi- 

 ties) is preferable for studies where only measurements of 

 Sr/Ca ratios are required. 



Materials and methods 



Otolith preparation 



Sagittal otoliths from an adult sockeye salmon (Oncorhyn- 

 chus nerka) collected in the Deschutes River, Oregon, and 

 a juvenile chinook salmon (O. tshawytscha) collected in 

 the Umatilla River, Oregon, were used to represent high 

 (>0.003) Sr/Ca and low (<0.001) Sr/Ca ratios, respectively 

 (Zimmerman, unpubl. data). High Sr/Ca ratios character- 

 ized the saltwater growth region in the sockeye salmon 

 otolith and low Sr/Ca ratios characterized the freshwater 

 growth region of the chinook salmon otolith. Each oto- 

 hth was mounted sulcus side down with thermo-setting 

 plastic resin on a microscope cover slip attached at one 

 end with super-glue to a standard microscope slide. The 

 otolith was then ground with 1200-grit sandpaper in the 

 sagittal plane to the level of the nucleus. The mounting 

 medium was heated and the otolith turned sulcus side-up. 



The otolith was then ground with 1200-grit and 2000-grit 

 sandpaper in the sagittal plane to the level of the primor- 

 dia and polished with 0.05-pm alumina paste. The cover 

 slip was then cut with a scribe and mounted with other 

 prepared otoliths (those used in other studies) on a petro- 

 graphic slide for microprobe analysis. The slide contain- 

 ing several otoliths was rinsed with deionized water, air 

 dried, and carbon coated (400 A). Elemental analysis was 

 conducted with a Cameca SX-50 wavelength dispersive 

 microprobe. Strontiantite (SrC03, USNM R10065) and 

 calcite (CaCOs, USNM 136321) were used as standards 

 for Sr and Ca, respectively. Standards were calibrated 

 with a 30-pm-diameter beam and 10-s counts resulting in 

 minimal effects of beam damage. 



Effect of spectrometer (crystal) choice 



To evaluate differences in diffracting crystals, we con- 

 ducted a series of tests where Sr was measured by using 

 both the PET and TAP crystals. A 15 kV, 50 nA beam was 

 used for these comparisons. With a 7-pm-diameter beam, 

 Sr was measured by using the TAP crystal (Sr La) and 

 Ca was measured by using the PET crystal (Ca Ka). Two 

 transects of 10 points each were sampled so that the points 

 on adjacent transects covered the same temporal location 

 on the otolith. Sr and Ca were analyzed simultaneously; 

 counting times for the Sr and Ca peaks were 40 s, and back- 

 ground counts were 40 s. A second set of transects covering 

 the same temporal locations in the otolith was sampled, but 

 Sr was measured with the PET crystal (Sr La). Because 

 our microprobe has only one PET crystal, simultaneous 

 measurement of elements was not possible. Transects were 

 conducted on both high and low Sr/Ca regions. Sr/Ca ratios 



