Santana and Lessa: Age and growth of Carcharhinus signatus off the northeastern Brazilian coast 157 



0° 



02°S - 



04°S - 



06°S 



ATLANTIC OCEAN 



Archipelago of 



Fernando de 



Noronha 



40 : W 



38=W 



36°W 



34 =W 



32W 



Figure 1 



Location of the sampling area for the night shark iC. signatus) collected off 

 northeastern Brazil. 



fin base to origin of the second dorsal fin [IDS, cm] ), total 

 length (snout to a perpendicular line from the tip of the up- 

 per caudal fin [TL, cm] ) and fork length (snout to fork of tail 

 [FL, cm]) were measured. In carcasses, only IDS was mea- 

 sured, and IDS, FL, and TL were recorded for eviscerated or 

 whole individuals. A set of five or six vertebrae were removed 

 from below the first dorsal fin in 317 specimens. Total length 

 was measured as the "natural length" (without depressing 

 the tail) according to Garrick ( 1982). 



To estimate TL for carcasses, relationships from sub- 

 samples of IDS versus TL and FL versus TL were estab- 

 lished for males and females separately. Linear regressions 

 derived for each sex were tested for homogeneity and ana- 

 lyzed for covariances (ANCOVA), resulting in TL=1.2049 

 FL + 1.7972 (r 2 =0.944.n=668,P=0.41) and TL = 3.3467 IDS 

 + 30.879 (r 2 =0.824; rc=764, P=0.161). Whenever length is 

 mentioned hereafter, we always refer to TL. 



Vertebrae were processed by removing excess tissue, 

 fixed in 49c formaldehyde for 24 hours, and preserved in 

 70% alcohol. Each vertebra was embedded in polyester resin 

 and the resulting block was cut to about a 1-mm thick sec- 

 tion containing the nucleus by using a Buehler® low speed 

 saw. Initially, alizarin-red-s stained sections (Gruber and 

 Stout, 1983) were compared to unstained sections from the 

 same individuals to define the best contrast for narrow and 

 broad zones. In the first procedure, sections were immersed 

 overnight in an aqueous solution of alizarin red s and 0.1% 

 NaOH at a ratio of 1:9 and then rinsed in running tap water. 

 In stained sections, narrow zones were visible as dark red 

 and broad zones as light red, whereas in unstained sections 

 translucent (narrow) and opaque (broad) zones were visible 

 under transmitted light. Unstained sections produced com- 



parable results to alizarin stained sections and were used 

 for band observation in the study. 



Bands counted in each section and distances from the focus 

 to the margin of each narrow zone were recorded. Vertebral 

 radius (VR) was measured by using a binocular dissecting 

 microscope equipped with an ocular micrometer. Measure- 

 ments were made at lOx magnification ( 1 micrometer unit=l 

 mm) with both reflected and transmitted light. The same 

 reader read sections from the same specimen twice at dif- 

 ferent times without knowledge of the individual size or 

 previous count. Whenever the counts differed between the 

 two readings, a third reading was used for back-calculation 

 of size-at-age. 



The index of average percentage error (IAPE) (Beamish 

 and Fournier, 1981) to compare reproducibility of age de- 

 termination between readings was calculated. 



IAPE = 1 / Ar]T ( 1 / R^ ( | X tj - Xj \Xj)x 100, 



where N = the number of fish aged; 

 R = the number of readings; 



X t - the mean age off 1 ' fish at the i' h reading; and 

 Xj = the mean age calculated for the/ ,! fish. 



Marginal increment ( MI ) analysis to determine the time 

 of band formation was used. The analysis was restricted to 

 1995-97, when samples were collected every month. The dis- 

 tance from the final band to the vertebral's edge (MI) was 

 expressed as a percentage of the distance between the last 

 two bands formed on vertebrae (Crabtree and Bullock, 1998). 

 The distance between the last and the penultimate band 

 was divided by the distance between the nucleus and the 



