Schmidt et al.: Spanish mackerel age, growth, and reproduction 



527 



of their range owing to their migratory nature, so speci- 

 mens were also acquired from recreational anglers, 

 commercial gill netters, and seafood dealers through- 

 out the southeastern United States. In addition, Span- 

 ish mackerel otoliths collected by the National Marine 

 Fisheries Service (NMFS) Panama City (Florida) Labo- 

 ratory were made available for this study. All speci- 

 mens used in the present study were collected between 

 Beaufort, North Carolina and Riviera Beach, Florida. 

 All length measurements refer to fork length (FL). Ex- 

 cept for those provided by NMFS, which were mea- 

 sured to the nearest cm, all fish were measured to the 

 nearest mm. 



One sagittal otolith from each fish was embedded in 

 paraffin and sectioned (ca. 0.3mm thick) with a Buehler 

 Isomet saw equipped with a diamond blade. Sections 

 were placed in cedar wood oil and examined under a 

 dissecting microscope with both transmitted and reflected 

 light. Most specimens provided by NMFS were processed 

 similarly, but some consisted of otolith sections mounted 

 on glass slides and were not immersed in cedar wood oil. 

 Ages were assigned by two independent readers without 

 reference to fish lengths or dates of capture. Otoliths 

 considered aberrant by either reader or for which the 

 readers disagreed on ages were deleted from the analy- 

 ses. Distances from the focus to the margin (otolith ra- 

 dius) and to the distal edge of the opaque portion of each 

 annulus were measured along the dorsal side 

 of the sulcus acousticus with an ocular mi- 

 crometer at 50 x magnification. 



Back-calculated lengths at age were com- 

 puted for males, females, and sexes com- 

 bined by the Fraser-Lee method (Poole, 

 1961; Carlander, 1982). The SYSTAT 

 NONLIN module using the quasi-newton 

 method (Wilkinson, 1987) was used to fit 

 von Bertalanffy equations to individual 

 back-calculated lengths at age. 



For all specimens except those from 

 NMFS {n = 1,742), the posterior portion of 

 the gonads was fixed in 10% seawater-for- 

 malin for 1-2 weeks and transferred to 50% 

 isopropanol for 1-2 weeks; samples from 

 NMFS (n = ni) consisted only of otolith sec- 

 tions with associated length and macro- 

 scopically determined sex (male vs. female) 

 data. Gonad samples were processed with 

 an Auto-Technicon 2A Tissue Processor, 

 vacuum infiltrated, and blocked in paraffin. 

 Three transverse sections (6-8 urn thick) 

 were cut from each sample with a rotary 

 microtome, mounted on glass slides, stained 

 with Harris hematoxylin, and counter- 

 stained with eosin-y. 



Sex and reproductive state were assessed by one 

 primary reader using histological criteria (Table 1). 

 Sections from 50 randomly selected specimens were 

 examined by a second reader early in the study to 

 verify the assessments of the primary reader. Because 

 only nine ripe female Spanish mackerel were captured, 

 the occurrences of developing and spent females were 

 also used to define the spawning period. Specimens 

 with developing, ripe, spent, or resting gonads were 

 considered sexually mature. For females, this defini- 

 tion of sexual maturity included specimens with oo- 

 cyte development at or beyond the yolk vesicle stage 

 and specimens with beta, gamma, or delta stages of 

 atresia. The SYSTAT NONLIN module (Wilkinson, 

 1987) was used to fit a logistic model (% mature=100%/ 

 1 + e- g "- L s»') to maturity data in 10 mm length inter- 

 vals to estimate length at 50% maturity (L 50 ). 



Because we were not able to obtain an adequate 

 number of female specimens near the beginning of the 

 spawning season to permit assessment of size/age at 

 maturity, we used specimens collected throughout the 

 year. To use all resting specimens in this assessment, 

 we developed a criterion to distinguish the small per- 

 centage (<10%) of resting females with no atresia from 

 immature (had never spawned) females. The criterion 

 was based on the size of chromatin nucleolar (CN; as 

 defined by Wallace and Selman 1981) oocytes, or oogo- 



