Bergenius et al Use of otolith morphology to indicate stock structure of Plectropomus leopardus 



501 



Doisal 



Ventral 



Figure 2 



Example of a common coral trout (Plectr'opoiniis leopardus) otolith reconstructed by using 

 lAi all of the 128 Fourier descriptors that were collected, and (Bi the first and last 14 

 descriptors. Otolith orientation and measures of length and width are indicated. 



B reef that was open for one year of fishing an(i then 

 closed, and one GU reef that was subjected to increased 

 fishing for a year prior to closure for five years. 



Because otoliths from only four-year-old fish were 

 analyzed in 1995 and 1999, the two cohorts included 

 individuals with nonoverlapping life histories. Conse- 

 quently, individuals collected in 1995 were not exposed 

 to the influence of the unusually large and persistent 

 Cyclone Justin that influenced the southern half of the 

 GBR throughout March in 1997, unlike those four-year- 

 old fish collected in 1999. Standardizing sampling by 

 age also minimized the potential for confounding spatial 

 variation in otolith shape with ontogenetic changes. 



Sagittal otoliths from up to 20 four-year-old P. leop- 

 ardiis were sampled from each reef each year (Table 1). 

 Otoliths from fewer than 20 fish were analyzed only if 

 less than this number were collected from a reef. This 

 sampling design enabled the examination of broad (re- 

 gion) and fine (reef within region) spatial and temporal 

 (1995 and 1999 cohorts) patterns in the otolith morph- 

 ology of P. leopa?-dus across much of the GBR, span- 



ning an unusual environmental event (Cyclone Justin, 

 1997) that had the potential to significantly influence 

 the results. 



Morphological analysis 



A microscope image (lOx magnification) was projected 

 onto a computer screen by using a video camera (Pana- 

 sonic GP-KR222E, Panasonic, Matsushita Communica- 

 tion Industrial Co., Osaka, Japan. Whole otolith area, 

 length, perimeter, and width (Fig. 2) and two shape 

 indices (circularity and rectangularity) were recorded 

 from each otolith by using the OPTIMAS image analy- 

 sis system (OPTIMAS, vers. 6.51, Silver Spring, MD). 

 Rectangularity was calculated as the area of the otolith 

 divided by the area of its minimum enclosing rectangle, 

 and circularity as the perimeter of the otolith squared 

 divided by its area. The perimeter of the otolith was 

 traced in a counter clockwise direction and digitized 

 into 128 x-y equidistant coordinates by using the distal 

 edge of the otolith rostrum as a common starting point 



