Pearson and Shaw: Age determination errors for Anoplopomo fimbria 



129 



OTC Mark 



Alignment point 



Alignment point 



Pasted UV Image 



Figure 2 



Composite image of a sablefish otolith. The otolith was first viewed under UV light 

 and an image was captured. It was then baked and a second image was captured by 

 using white light. Then a small rectangle from the UV image was electronically cut 

 and pasted on the image of the baked otolith. The fiourescent mark produced by the 

 OTC appears as a dark line on the UV section. Points on the otolith used for correct 

 positioning of the pasted section are shown. 



The otoliths were embedded in epoxy casting resin. After 

 the resin hardened, the blocks containing the otoliths were 

 sliced in half across the dorsoventral axis with a diamond 

 saw. 



Images were captured in a two-stage process. The first 

 stage used ultraviolet light to reveal the OTC mark, and the 

 second stage used white light to reveal the growth marks 

 used for age determination. In the first stage, the room was 

 completely darkened and an image of the otolith, including 

 the OTC mark, was captured by using a video camera capa- 

 ble of capturing images under low light conditions. We used 

 an ultraviolet lamp which produced a strong beam of light 

 at 365 angstroms. The otolith was viewed on a compound 

 microscope using reflected light. The camera and image pro- 

 cessing system were connected to a PC computer equipped 

 with a frame grabber card. A version of NIH Image, a pub- 

 lic domain image processing software (Scion Corporation, 

 Frederick, MD), was used to process the images. 



The embedded otolith was placed on the microscope and 

 a drop of mineral oil was placed on the surface of the oto- 

 lith. The limited amount of UV light available to the cam- 

 era required the use of frame averaging. Usually 30 frames 

 were sufficient to produce a sharp view of the otolith and 

 the fluorescing mark. In some cases, the mark was too faint 

 to allow an image to be captured. When there was sufficient 

 fluorescence, two composite images were captured, one at 

 4x and one at 40x. 



In the second stage, the same embedded otolith was 

 placed in a small toaster oven at 270°C and heated for 20 



to 25 minutes until it had turned dark brown. This baking 

 process enhanced the growth rings for visual analysis and 

 approximated what age readers see using the break and 

 burn method; however, the latter process results in darker 

 hyaline zones than those obtained with this method. After 

 cooling, the otolith was viewed under white light. A second 

 set of images was then captured. A section of each UV im- 

 age was then electronically cut and pasted onto the image 

 captured under visible light. With some experimentation it 

 was found that the pasted sections could be aligned exactly 

 over the visible light images, creating a final composite im- 

 age as shown in Figure 2. 



Initial examination of the otoliths 



Initially, all OTC-marked otoliths were examined with 

 knowledge of the year and season of release, but without 

 any other information about the fish. Composite UV and 

 white light images were obtained as previously described. 

 The age reader determined the following: whether or not 

 the OTC mark was visible; whether the OTC mark was in 

 a hyaline or opaque zone; the number of annual hyaline 

 zones visible beyond the OTC mark (and whether or not 

 the edge was included in the count); edge type (hyaline, 

 narrow opaque, wide opaque, or unidentifiable), and the 

 shape of the otolith. In some cases the OTC mark could 

 not be identified or the mark was too faint to be captured 

 as a composite image; these specimens were excluded from 

 subsequent analyses. 



