548 
Fishery Bulletin 95(3), 1997 
& 
o 
-16 
•17 
-18 
-19 
-20 
10 
1 1 
12 
13 
1 4 
15 
5 15 N (%o) 
Figure 5 
Weighted mean 6 15 N versus 8 13 C values for all king mackerel sample collection sites 
and regions. Number of samples per site can be seen in Table 2. Standard error bars 
are shown for sites. Regions were defined according to stable isotopic data and stock 
structure studies (Baughman, 1941; Fry, 1983; Macko et al., 1984; Trent et ah, 1987; 
Fable et ah, 1990; Johnson et ah, 1994; May 2 ). 
ary water masses off Mexico 
and south Texas may collide 
and form a convergence zone 
that directs low-salinity wa- 
ters offshore, near Browns- 
ville, TX (Vastano 5 * ) (Fig. 1). 
This convergence zone may 
act as a temporary boundary 
between northwestern Gulf 
of Mexico fish and Mexico 
fish. 
The innate migratory pat- 
terns of the king mackerel 
can influence the isotopic val- 
ues observed in their dorsal 
spines. The location in which 
food is assimilated should 
directly influence the isotopic 
value recorded in the spine. 
Consequently, the area in 
which the individual fed, 
rather than collection site of 
the individual, would be de- 
tected in the spine. There- 
fore, determination of groups 
of king mackerel from collec- 
tion site alone, may be inap- 
propriate. In addition, re- 
gional groupings of the mack- 
erel that were based upon 
isotopic data and on previous king mackerel studies 
may be more suitable for drawing conclusions. 
Likewise, the migratory nature of king mackerel 
complicates the use of season of collection as a vari- 
able (Table 2). For example, the GLM (Eq. 3) indicated 
that the season in which the specimens were collected, 
fork length, and collection site influenced the nitrogen 
isotope results. However, when the data were divided 
into regions, fork length and region were the only vari- 
ables that had a significant effect on the GLM (Eq. 4). 
Again, consideration of the data by region, as opposed 
to individual site, may be more appropriate, particu- 
larly because, in the former case, time of collection did 
not bias the isotopic findings. 
Our method of using stable isotopes is a new ap- 
proach in trying to determine the number and loca- 
tion of king mackerel groups. An advantage of isoto- 
pic analysis over that of genetics is that stable iso- 
topes enable researchers to view significant changes 
in an individual, whereas genetic methods require 
generations to see significant variations. The disad- 
vantage to stable isotopes is that the signal is ac- 
5 Vastano, D 1995. Oceanography Department, Texas A&M 
Univ., College Station, TX 77843. 
quired only from areas in which food is assimilated, 
which may or may not represent the location and 
number of king mackerel groups. Although the num- 
ber and location of king mackerel stocks have been 
researched previously by using genetic techniques 
(see below), several scenarios exist. Research by 
DeVries and Grimes (1991) has suggested the possi- 
bility of three stocks: a western Gulf of Mexico, an 
eastern Gulf of Mexico, and an Atlantic stock. From 
mitochondrial DNA data, Gold et al. (in press) found 
weak genetic differences between Atlantic and Gulf 
of Mexico king mackerel that implied more than one 
stock. Additionally, Johnson et al. (1994), using elec- 
trophoretic data, suggested the existence of two 
stocks, eastern and western, within the Gulf of 
Mexico. The idea of separate eastern and western 
stocks of king mackerel within the Gulf of Mexico 
has also been supported by Baughman ( 1941), May, 2 
and Trent et al. (1987) with observational, electro- 
phoretic, and catch results, respectively. 
Our 8 15 N data showed significant differences be- 
tween king mackerel caught in Mexican and Florida 
waters in contrast to those collected in the north- 
western Gulf of Mexico. Thus, our isotopic results 
suggest that at least two distinct groups exist within 
