FISHERY BULLETIN: VOL. 84, NO. 3 



grow very slowly after sexual maturity (sexual 

 maturity for female Atlantic sailfish reported at 

 13-18 kg, Jolley 1977). For example, tagging records 

 indicate that this fish, which was tagged at 18.2 kg, 

 gained only about 6.4 (14 lb) while being at-large 

 almost 1 1 yr. Our analysis of spines and otoliths sup- 

 port these findings. 



Dorsal Spines 



Examination of sections from dorsal spines 3-6 

 (Fig. 2) indicated that the vascularized core com- 

 prised an extensive area in all sections. The solid 

 bone area where zonations were not disrupted 

 varied in size and comprised 14, 19, 30, and 37% 

 of the right lobe of spine sections 3, 4, 5, and 6, 

 respectively (Table 1). The vascularized core severe- 

 ly restricted the zonation counts because increments 

 associated with early growth history were totally 

 disrupted and could not be enumerated. Counts of 

 zonations on the four spine sections ranged from 2 

 to 8 (Table 1). This suggests that spine number 4, 

 which had been used in past studies to assign ages, 

 may not necessarily be the best choice for ageing 

 sailfish, particularly for the larger, older specimens. 

 For example, spines 5 and 6 both had a higher per- 

 centage of solid bone, and counts of zonations in 

 these spines were proportionately higher than in 

 spines 3 and 4 (Fig. 2, Table 1). However, all spines 

 substantially underestimated the age of this sailfish, 

 where approximate age (13-15+ yr) was known 

 from tagging records. Hedgepeth 6 reports that spine 

 number 4 would not have been included in the data 

 sets of previous published studies because of the ex- 

 tensive size of the vascularized core area. We con- 

 clude from these data that dorsal spine sections are 

 probably only useful for ageing sailfish from > 1 to 



6 Hedgepeth, M. Y., Fisheries Biologist, Florida Department of 

 Natural Resources, 727 Belvedere Rd., West Palm Beach, FL 

 33405, pers. commun. 1984. 



Table 1 .—Mean count of zonations (3 repetitions) and percentage 

 solid bone in the distal portion of the right lobe of sections taken 

 from dorsal spines 3-6 of Atlantic sailfish (see text and Fig. 2). 

 Measurements and counts were taken along a straight line count- 

 ing path bisecting the spine laterally from the focus to the edge 

 of each section. 



5 yr. Although there may be some bias associated 

 with ageing these young sailfish because of the 

 vascularized core, this bias is probably minimal. 

 However, for sailfish older than estimated age 5 and 

 about >22.7 kg (50 lb), the bias substantially under- 

 estimates age and this bias increases with an in- 

 crease in size and age of the fish. In addition, spines 

 have not been shown to be useful in ageing sailfish 

 <l-yr-old (Jolley 1974, 1977). 



Otoliths 



Sagittae from the tagged Atlantic sailfish had ex- 

 ternal and internal morphologies that were charac- 

 teristic of sailfish reported by Radtke and Dean 

 (1981), as well as other istiophorids (Radtke et al. 

 1982; Wilson and Dean 1983; Radtke 1983). For ex- 

 ample, major features of these sagittae include a 

 rostrum and antirostrum separated by a deep sulcus 

 (Fig. 3). The external ventral and lateral surfaces 

 of the rostrum consist of a series of ridges that are 

 perpendicular to the axis of growth (Fig. 4). Radtke 

 and Dean (1981) suggested that the number of 

 rostral ridges can be used to estimate age of Atlan- 

 tic sailfish. To make an accurate count of external 

 ridges for age estimation, it is necessary to under- 

 stand the internal and external otolith growth pat- 

 tern so that the location and number of the first few 

 rostral ridges can be firmly established. These ini- 

 tial ridges are often covered by excess calcium car- 

 bonate (Wilson 1984), particularly in older speci- 

 mens, and are not always visible on the external 

 features of the lateral surface (Fig. 4). 



The growth of the rostrum occurs in two direc- 

 tions (Figs. 3, 4). During early stages, incremental 

 growth of the rostral lobe occurs in the ventral direc- 

 tion out to a bend where growth shifts to a more 

 medioventral and then to a medial direction (Fig. 

 3). This same pattern of otolith growth has been 

 reported for blue marlin, Makaira nigricans, and 

 white marlin, Tetrapturus albidus (Wilson 1984). 

 However, it is difficult to illustrate a complex three- 

 dimensional otolith on a two-dimensional photo- 

 graph. Therefore, Figures 3 and 4 should be ex- 

 amined consecutively to obtain a proper orientation 

 of the structure. 



Although rostal ridges on the external lateral sur- 

 face (Fig. 4) are not distinct because of the excess 

 calcium carbonate, after the change in the axis of 

 growth, the ridges on the ventral surface (ridges 

 3-10) can be counted easily (Fig. 4). Several lines of 

 evidence points towards the first two growth zones 

 occurring within the boundaries of the lateral sur- 

 face. For example, a distinct internal translucent 



496 



