FISHERY BULLETIN: VOL. 84, NO. 3 



zone exists between the boundary of many of the 

 external ridges shown in Figure 3. This zone extends 

 from the surface to deep within the internal struc- 

 ture of the rostrum (Fig. 3). The distinct change in 

 optical density of the first of these prominent zones 

 marks what we believe is the boundary between the 

 1st and 2d ridges and suggests that the first major 

 growth zone is located between the core and the 

 bend (Fig. 3). Further, using the SEM we counted 

 150-200 finely spaced increments between the core 

 and the first prominent translucent zone. This count 

 also supports our interpretation of the location of 

 the first annual zone if these increments are as- 

 sumed to form daily, the fish was born sometime 

 in late spring or early summer (as reported by 

 Beardsley et al. 1975), and the annual zonations are 

 being formed in the winter. Jolley (1977) reported 

 that annual zones in Atlantic sailfish spines tend to 

 be formed in late fall or winter and he also spec- 

 ulated that sailfish may form the first annuli on 

 spines prior to a full year's growth. The location of 

 the second translucent zone, based on similar evi- 

 dence, appears to be at the beginning of the bend 

 (Fig. 3). The width of the first two major growth 

 zones (^0.5 mm) are considerably larger than the 

 zones beyond the end. Wide spacing of year marks 

 during early growth have been observed in many 

 fishes when growth rates are most rapid (Dean et 

 al. 1983). Therefore, these data support our conten- 

 tion that at least two ridges should be accounted for 

 as occurring within the boundaries of the lateral 

 surface. 



Rostral ridges 3 through 10 were easily distin- 

 guished and counted on the sagitta's ventral surface 

 within the same plane of focus (Fig. 5, bottom). 

 After the 10th ridge, however, the rostrum changes 

 direction slightly (Fig. 3), and it was necessary to 

 refocus to observe ridges 11 through 13 (Fig. 5, top). 

 We feel that potential sources of error in our counts 

 of rostral ridges would have most likely occurred 

 at the beginning and end of the counting path. In 

 addition, we feel that if errors were made at these 

 locations, they would have increased the count. 

 Therefore, otolith age of the tagged Atlantic sailfish 

 was estimated to be 13 yr. However, it should be 

 recognized that potential errors in this estimate 

 could have resulted if one or two ridges were un- 

 accounted for on the lateral surface or on the tip 

 of the rostrum on the ventral surface. Otolith age 

 under these circumstances should be presented con- 

 servatively as ranging from 13 to 15+ yr. 



The weight of one sagitta from the tagged Atlan- 

 tic sailfish (7.84 mg) was extremely heavy for an 

 istiophorid of comparable size. For example, it was 



1.24 mg heavier than the sagitta from a 29.6 kg (65 

 lb) sailfish caught in 1985 off Miami and was 1.18 

 mg heavier than the largest sagitta from Pacific blue 

 marlin reported by Radtke (1983). In addition, the 

 tagged sailfish sagitta was in the upper range in 

 weight (0.51-8.16 mg) of more than 500 blue and 

 white marlin sagittae examined by Wilson (1984). 

 Since the relationship between the size of otoliths 

 and the age of fishes has been shown to be positive- 

 ly correlated for some teleosts (Somerton 1985), we 

 feel that the relatively large size of this sagitta pro- 

 vides additional indirect evidence that this structure 

 could be from a very old sailfish. 



CONCLUSIONS 



Our tagging records indicate that estimates of 

 maximum longevity for Atlantic sailfish should be 

 revised upwards to at least 13-15+ yr, and that 

 sailfish of this age can grow at a very slow rate 

 (about 0.59 kg/yr during its time at large). Dorsal 

 spines do not appear to be an accurate source of age 

 and growth information for older, larger sailfish (>5 

 yr, ^22.7 kg or 50 lb), while sagittae do provide more 

 accurate estimates of age for these older age groups. 

 Since current stock assessments of Atlantic sailfish 

 (Conser 1984) rely exclusively on dorsal spine ageing 

 data as input, these assessments offer little insight 

 into the more mature segments of the population. 

 If skeletal structures from the larger, older fish are 

 systematically rejected for ageing analyses, an 

 underestimate of age and longevity and an 

 overestimate of growth rate can occur (Nammack 

 et al. 1985). Therefore, future assessments should 

 be revised using otolith ageing methods to clarify 

 that portion of the age structure that can not be 

 reliably appraised using dorsal spines. 



ACKNOWLEDGMENTS 



We thank J. T. Reese Taxidermist, Inc., Ron Har- 

 rison (angler), and Captain Bud Carr for providing 

 us with biological samples and other information 

 from the tagged Atlantic sailfish. Personnel from 

 the Florida Department of Natural Resources in 

 West Palm Beach, FL, sectioned and analyzed dor- 

 sal spine number 4. Dana Dunkleberger (University 

 of South Carolina) assisted in preparing scanning 

 electron micrographs. 



LITERATURE CITED 



Beardsley, G. L. 



1980. Size and possible origin of sailfish, Istiophonts 



500 



