96 



Fishery Bulletin 97(1), 1999 



100 



80 



60 



S 40 



20 



P=99 89/{l+[(99.89-0 00024)/0 00024][exp(-0.274TL)]} 



n=31 



r=0 998 



36 38 40 42 44 46 48 50 52 54 56 58 60 62 64 66 



Total length (cm) 



Figure 4 



Sexual maturity o^ve for P. prometheus caught off the Canary Islands (August 1992-July 

 1995). 



spawning of this species occurs from August to Sep- 

 tember in the Atlantic off Madeira. Nishikawa ( 1987 ) 

 found the highest concentration of gempyHd larvae 

 at the end of the summer months in the Pacific wa- 

 ters of Japan. In the Canary archipelago, the spawn- 

 ing of roudi escolar seems to be related to water tem- 

 peratures, occurring when these reach greatest val- 

 ues. Possible significance of seasonal temperature 

 variation to maturation and spawning in other bony 

 fish species off the Canary Islands has been discussed 

 by Lorenzo and Pajuelo (1996) and Pajuelo and 

 Lorenzo (1995, 1996). Reproduction in roudi escolar 

 does not involve horizontal migrations because dur- 

 ing the spawning season the specimens are observed 

 in the same areas where they are fished all year 

 round. The lowest discrepancy in the number of males 

 and females obser\-ed during the spawning period 

 suggests an aggregation for breeding. Males and fe- 

 males were found aggregated mainly at depths of 

 450-650 m during this period. 



Length at sexual maturity does not differ between 

 males and females, corresponds approximately to 48 

 cm TL. In the age-length relationship, this size cor- 

 responds to 4-year-old specimens. The size at which 

 fifty percent of the fish become mature is less than 

 the length at first capture and the majority of the 

 total catch is longer than this length, indicating a 

 good exploitation pattern from the biological point 

 of view. Furthermore, a low value of fishing mortal- 

 ity rate was obtained. 



The alternate pattern of opaque with translucent 

 zones was easily distinguishable on the otoliths of 

 the roudi escolar. These zones are deposited owing 

 to alternating periods of rapid and slow growth (Wil- 

 liams and Bedford, 1974). The opaque zone is formed 

 when the water temperature is higher, and food is 

 abundant, and the translucent is formed when tem- 

 perature is lower and the species spawns. This find- 

 ing demonstrated the validity of using otoliths for 

 estimating the age and growth of roudi escolar. The 

 false hyaline zones observed within the fourth opaque 

 ring and in subsequent opaque rings are probably 

 spawning bands, because this species spawns in the 

 summer months, when the opaque zone is formed in 

 the otoliths. Morales-Nin (1987) pointed out that 

 when the spawning does not take place during the 

 period of hyaline zone formation, a false ring, known 



