422 



Fishery Bulletin 103(2) 



method), these estimates have to be seen only as indica- 

 tions that the asymptotic length of E. oeulatus is quite 

 large and that the Hondurian population is moderately 

 exploited (if M = 2K, as suggested by Ralston (1987) for 

 snappers, then ZIK = 3.73 and E = F/Z = 0.46). 



Reproduction 



Macroscopic observations of gonads were recorded for 

 309 fish whose sex could be identified (118 males and 

 191 females); all stages of the reproductive cycle were 

 observed, but only 20 individuals were in the prespawn- 

 ing to postspawning stages, and a single one was found 

 to be in the process of spawning. 



The smallest fish with developing gonads was 36 cm 

 for females and 29 cm for males (Fig. 4). Although 

 only part of the length range of males was adequately 

 sampled (100 out of the 118 fish were smaller than 44 

 cm I, it appears that the progressive build-up of the 

 reproductive male population occurs between about 30 

 cm and 45 cm. The picture is clearer for females, whose 

 sample size was larger and more evenly spread over 

 the length range: above 54 cm, all females were found 

 to be in a reproductive cycle. The maturing process 

 therefore occurs at clearly lower sizes for males (30-45 

 cm) than for females (35-55 cm). Females in advanced 

 reproductive stages (postspawning and resting stages) 

 were observed across the length range, including the 

 smallest adult sizes, those below 45 cm (Fig. 4). 



A full analysis of the seasonality of reproduction is 

 not possible because data were collected in only seven 

 months, out of which only four (May, June, November, 

 and December) yielded samples large enough for the 

 analysis (21 to 72 females per month). No females were 

 found to be spawning, but most of the pre- and post- 

 spawning stages (14 out of 17) were observed in Novem- 

 ber and December, and half of maturing females were 

 fished during the last quarter of the year (Fig. 5). How- 

 ever, 74% of females at sexual rest (resting stage) were 

 caught in May and June. Additional pieces of informa- 

 tion confirm this pattern: the only spawning individual, 

 a male, was observed in November (Dominica); females 

 gonads in advanced stage of vitellogenesis were observed 

 in September (Guadeloupe); no mature individual was 

 found in Honduras in April-June. These observations 

 show that an active spawning period occurs at the end 

 of the year (even if all fish caught at this period were 



not close to the spawning phase), as opposed to late 

 spring which is a period of sexual inactivity. 



Such limited results leave open the overall interpre- 

 tation of the annual reproductive cycle of E. oeulatus. 

 In particular, according to the fishermen working on 

 the SMSBA shelf, the species could have an extended 

 spawning season, lasting from November to April or 

 May. 



Predators and prey 



No systematic observations were made on the trophic 

 relationships of E. oeulatus, but a few occasional record- 

 ings were made of its predators and prey. The only record 

 of a predator was that of a beardfish {Polymixia lowei: 

 Polymixiidae) measuring 40 cm TL containing a very 

 small queen snapper (8.5 cm TL) and which was caught 

 deeper than 200 m. This is the first record of such a food 

 item for this beryciform fish whose diet had so far been 

 reported to comprise cephalopods (Cervigon, 1991). The 

 stomachs of E. oeulatus that could be observed were most 

 often empty; on a few occasions, unidentified squids were 

 the only items present. This was the case for three fish 

 (58 to 62 cm) caught at 430 m depth. 



Discussion 



Etelis oeulatus was found on the upper part of the con- 

 tinental and insular slopes, from about 150 to 450 m; 

 this observed range confirms previous indications (Allen, 

 1985), but the bathymetric distribution of the species 

 could possibly extend beyond the maximum depth fished 

 in these surveys. The presence of E. oeulatus in shal- 

 lower waters of the shelf seems possible, according to a 

 statement that juveniles can be found in less than 30 m 

 (Appeldoorn et al., 1987) and to the reported catch of one 

 fish (size not recorded) at 59 m depth by a trawl survey 

 off southeastern United States (Cuellar et al., 1996). 

 However the present data, other fishery-independent 

 surveys focusing on snappers (i.e., Marcano et al., 1996, 

 down to 128 m), and most studies on Caribbean coastal 

 fisheries strongly indicate that the species is very rare 

 on the shelf itself. 



Within the observed depth range, there is a ten- 

 dency for the largest fish to be found in the deeper 

 areas, as observed in the closely related Pacific species 



