322 



Fishery Bulletin 89(2), 1991 



released as southeastern Florida winter substock were 

 also classified into all substocks, yielding a 70.60% 

 affinity-index value (combined Gulf, 0.43%; Florida 

 Keys winter, 55.36%; and southeastern Florida winter, 

 14.81%). 



Discussion '• 



Two movement patterns were identified from our tag- 

 ging study. King mackerel released in association with 

 the Gulf of Mexico waters were found during winter 

 months along the Florida Keys and along the south- 

 eastern Florida coast as far north as Cape Canaveral. 

 By spring, these fish traveled along the western Florida 

 coast toward northeastern Gulf waters, continuing 

 westward during the summer. These king mackerel 

 returned toward northwestern Florida in late-summer 

 and early-fall, and then headed back to southern 

 Florida waters by winter. Based on the recapture of 

 fish from the same locations during roughly the same 

 times of year over several consecutive years (e.g., Fig. 

 4a-d), we conclude that a periodic (annual) migratory 

 behavior exists. The regularity of king mackerel move- 

 ments through Gulf of Mexico waters was also noted 

 during other tagging studies (Sutherland and Fable 

 1980, Fable 1988). 



Trends of the catch-per-unit-effort (CPUE) for king 

 mackerel from charterboat catches, reported by Trent 

 et al. (1987), also support our conclusions. During 1983, 

 1984, and 1985, CPUE values of king mackerel gener- 

 ally increased along the northern Florida coast during 

 late-May and June, followed by a peak in CPUE along 

 the Alabama, Mississippi, and Texas coastlines. An- 

 other increase in CPUE was noted in Alabama during 

 late-July and August, followed by a peak in northwest- 

 ern Florida during late-August or early-September. 

 Only winter peaks were noted in southern Florida, 

 whereas no CPUE pattern was established for south- 

 eastern Florida. The lack of a consistent peak in this 

 region as determined by tag returns from southeastern 

 Florida may be evidence for a resident (i.e., non- 

 migratory) population. Fable et al. (1987) noted winter 

 resident population of large king mackerel (>800mm 

 FL) in the northwestern Gulf of Mexico. We found no 

 size-component associated with tagged fish either stay- 

 ing or leaving southeastern Florida waters (x 2 8.19, 

 P>0.975, df 18), based on length-frequency distribu- 

 tions of recaptured fish (50-mm FL interval, 400-1300 

 mm). 



Movement of king mackerel along the Atlantic coast 

 was not as clearly defined. Fish traveled south from 

 South Carolina waters during the spring and summer, 

 distributing themselves along the south Atlantic coast 

 as far as the eastern and southwestern coasts of south- 



ern Florida, and then returned northward in late- 

 summer and early-fall. Trent et al. (1987) noted an in- 

 crease in CPUE from charterboat catches from North 

 Carolina during April and again during October and 

 November. Progressive increases in CPUE were noted 

 in South Carolina during May and later during August 

 and September; peaks were noted in Georgia and 

 northeastern Florida during late-May (after South 

 Carolina peaks) and again during August (before South 

 Carolina peaks). 



Spring and summer movements of king mackerel 

 may reflect migrations to their respective spawning 

 areas. After spawning, they return to wintering areas 

 in the fall. Ichthyoplankton collections indicate that 

 king mackerel from the Atlantic coast spawn from 

 April through October, with a peak during September 

 (Collins and Stender 1987). Gulf of Mexico larvae col- 

 lections and reproduction indices suggest a similar 

 spawning season for the Gulf stock (Wollam 1970, 

 Dwinell and Futch 1973, McEachran et al. 1980, Finu- 

 cane et al. 1986). 



King mackerel movement patterns, as determined 

 from temporal and spatial variability in tag returns, 

 must be viewed in relative and descriptive terms rather 

 than in absolute rates of movement. The percentages 

 of returns varied from 5.69% (South Carolina region) 

 to 9.68% (Florida Keys region), but when numbers of 

 recaptures were compartmentalized by subareas, these 

 values were subsequently reduced. This reduction was 

 magnified as distance from release site increased. 

 Lower percentage of returns may be a function of ac- 

 tual proportion of fish moving away from location of 

 release but may also reflect relative changes in effort 

 and availability over time, or may be indicative of a 

 "dilution" of tagged fish by king mackerel from other 

 locations (e.g., with a resident population in the north- 

 western Gulf of Mexico). Combinations of these factors, 

 as well as other environmental and biological 

 parameters, are probably interacting, but are beyond 

 the resolution ability of our database. However, we feel 

 the observed trends, together with published informa- 

 tion previously discussed, are strong enough to support 

 our conclusions of king mackerel movement through 

 SE U.S. waters. 



Distance from the 'transition zone' along the south- 

 ern Florida coast (Fig. 1, shaded area) was related to 

 the affinity-index values calculated for each substock. 

 Both the South Atlantic and combined Gulf substocks, 

 located outside this transition area, had 100% classifica- 

 tion (affinity) values with the appropriate Atlantic or 

 Gulf stock. The Florida Keys winter substock, located 

 in the southwest edge of this transition area, was 

 closely tied to the Gulf stock (90.00% affinity). King 

 mackerel released in southeastern Florida had the 

 lowest affinity values (Table 3). Winter releases from 



