Roelke and Cifuentes: Use of isotopes to assess groups of Scomberomorus cavalla 
541 
±1.1 %c occurs for nitrogen isotopes (Minagawa and 
Wada, 1984). Macko et al. (1984) also demonstrated 
that stepwise enrichment, which occurs from trophic 
level to trophic level, does not vary among locations. 
Finally, Minagawa and Wada (1984) suggested that 
individuals do not fractionate nitrogen isotopes dif- 
ferently at various ages. 
Estep and Vigg (1985) observed that the carbon 
isotope discrimination between scale and muscle was 
consistent for a particular fish species, and stated 
that isotopic measurements in muscle and scales 
could be used to determine diet of fish. Studies that 
encompass time scales of months to years, however, 
may be compromised by fast turnover of muscle. 
Collagen, in contrast to muscle, has a slow turnover 
rate of carbon (Libby et al., 1964). Collagen amino 
acid composition varies only slightly among species 
(see Schoeninger and DeNiro, 1984); therefore, dif- 
ferences in collagen isotope ratios reflect isotopic 
changes in diet and not variations in chemical com- 
position (Schoeninger and DeNiro, 1984). Addition- 
ally, we speculate that turnover of collagen may be 
slower in poikilotherms, such as fish, compared with 
homeotherms, such as mammals, owing to the lower 
metabolic rate of poikilotherms. The chemical uni- 
formity and slow turnover time of collagen make it a 
suitable matrix for recording the dietary history of or- 
ganisms that grow fin spines (fin spines consist of col- 
lagen fibers in a bony matrix) and live over periods of 
years. 
In the Gulf of Mexico, Fry (1983) compared stable 
carbon (8 13 C) and nitrogen (5 15 N) isotope ratios of 
several decapod crustaceans and two species of in- 
shore fishes. The study revealed that groups feed- 
ing primarily in the eastern Gulf of Mexico have 
different 8 15 N and 8 13 C values from those feeding 
in the western Gulf of Mexico (Table 1). Macko et 
al. (1984) also reported geographical variations in 
isotopic ratios of sedimentary organic matter. From 
these studies, we hypothesized that stable carbon 
and nitrogen isotopes would aid in establishing 
group structure if 1) king mackerel feed in differ- 
ent regions of the Gulf of Mexico for extended peri- 
ods of time, and 2) the carbon and nitrogen isotope 
composition of a food source is incorporated in a 
consistent manner into tissues of king mackerel. 
Dorsal fin spines were chosen for isotopic analyses 
because collagen has a slow turnover rate, and the 
isotopic ratio of the spine should reflect assimilated 
food at the time of formation. Therefore, the diet 
recorded within the spine is chiefly a record of early 
developmental years, when the majority of growth 
occurs. A previous study, however, showed a loss of 
the first annulus in fin spines of older swordfish 
(Tsimenides and Tserpes, 1989). We could find no 
such study of fin spine annulus of king mackerel to 
substantiate this loss in king mackerel. Loss of mass 
has not been investigated in swordfish research; there- 
fore, we assumed that if any material is lost, it is mini- 
mal in comparison with the remaining material, be- 
cause this phenomenon has been observed only in 
large individuals. Below, we report on the identifica- 
tion of two isotopically distinct groups of king mack- 
erel in the Gulf of Mexico and compare our findings 
with previous studies conducted in order to determine 
location and number of king mackerel groups. 
Table 1 
Mean stable isotope (C and N) values of sediment, particulate organic matter, zooplankton, shrimp, and mackerel for Florida, 
Northwestern Gulf of Mexico, and Mexico. Standard error is presented when available. GOM = Gulf of Mexico, nd = no data. 
Florida king mackerel data comprise collection sites Panama City, FL, and Fort Pierce-Palm Beach, FL. Northwestern king 
mackerel data comprise collection sites Port Aransas, TX; Galveston, TX; Grand Isle, LA; and GulfPort, MS. Mexico king mack- 
erel data comprise collection sites Dzilam DeBravo, Celestun, and Veracruz. 
Sample type 
5 13 C 
5 15 N 
Florida 
Northwestern GOM 
Mexico 
Florida 
Northwestern GOM 
Mexico 
Sediment 7 
-18.5 ±0.7 
-20.610.6 
nd 
3.610.1 
6.510.2 
nd 
Particulate organic matter 7 
-19.4+1.2 
-21.0 ±1.4 
nd 
-0.9 ±1.4 
7.510.8 
nd 
Zooplankton 7 
-18.411.1 
-19.2 ±0.7 
nd 
5.9 10.7 
8.9 10.9 
nd 
Penaeus shrimp 7 
-14.810.5 
-15.611.1 
nd 
8.4 ±0.9 
12.911.1 
nd 
Penaeus shrimp 2 
-14.6 
-15.9 
nd 
8.3 
12.6 
nd 
King mackerel (this study) 
-18.911.1 
-18.1 ±0.9 
-17.711.5 
10.8 ±1.1 
13.1 ±1.3 
10.8 11.0 
7 Data, excluding the king mackerel data, were compiled from Fry (1983 land Macko et al. (1984). 
2 Data are based on estimated values from Figure 7 in Fry (1983). 
