166 
Fishery Bulletin 107(2) 
ported as a ratio of heavy to light isotopes and given 
d notation with units of %o, see Materials and methods 
section for more detail) have been used to determine 
the importance of MDN in freshwater systems, and to 
characterize the trophic structure within those systems 
(Kline, et al., 1993; Vander-Zanden et al., 1999). For 
example, carbon and nitrogen isotopes have shown that 
anadromous Pacific salmon ( Oncorhynchus spp.) were a 
significant source of allochthonous nitrogen to coastal 
streams where spawning occurs (Kline et al., 1993). 
Hesslein et al. (1991) used sulfur isotopes to differen- 
tiate freshwater migratory and non-migratory fishes in 
the Mackenzie River Basin, Canada. On the East Coast 
of the United States, anadromous river herring ( Alosa 
spp.) retain their marine isotope signal after spending 
part of the spring spawning in freshwater, and that 
some freshwater piscivores are 34 S and 13 C-enriched 
after preferentially consuming migrating Alosa spp. 
during the spawning run (Garman and Macko, 1998; 
MacAvoy et al., 2000). 
An additional tool for determining origins and trans- 
formations of organic material from different sources 
is the stable isotope ratio of specific compounds. Isolat- 
ing a specific compound, or class of compounds, then 
measuring the isotope ratio on those compounds, may 
offer a more robust technique to trace biologically 
significant compounds (such as fatty or amino acids) 
than would be possible from bulk isotope analysis 
alone. For example, examining the carbon isotopic 
composition of fatty acids from an animal, particularly 
essential fatty acids, allows the direct determination of 
dietary sources that contribute to the fatty acid pool of 
that animal (Stott et al., 1997). Although bulk isotope 
analysis can be an effective nutrient tracer in systems 
with isotopically distinct nutrient sources (Peterson 
et al., 1985), the isotopes of specific fatty acids may 
provide more confidence in identifying sources (Canuel 
et al., 1997). 
Carnivorous heterotrophs are unable to synthesize 
fatty acids longer than 18-carbons, nor can they de- 
saturate carbon-carbon bonds between the ninth and 
terminal methyl carbon, therefore, these essential fatty 
acids must be obtained from diet (Olsen 1999). Because 
essential fatty acids are not influenced by subsequent 
metabolism within a eukaryotic heterotroph, they re- 
tain their original isotope composition (Stott et al., 
1997). Fatty acids synthesized by marine plankton 
and incorporated into marine fish would be highly en- 
riched in 13 C relative to those produced by freshwater 
primary producers or C3 photosynthesis. Addition- 
ally, short chain fatty acids, used as precursors in the 
biosynthesis of unsaturated or longer chain saturated 
fatty acids, should be 13 C enriched in relation to bio- 
synthesized fatty acid products (Murphy and Abrajano, 
1994). In this study, the fatty acid nomenclature used 
is carbon numberinumber of double bonds. For ex- 
ample, 18:2 is an 18-carbon fatty acid with two points 
of unsaturation. The desaturation of 16:0 to 16:1 and 
18:0 to 18:1-18:2 occurs by a systematic fractionation 
of roughly 2 %c per desaturation (DeNiro and Epstein, 
1977; Monson and Hayes, 1982). Also, studies have 
shown that the elongation of fatty acids by de novo 
synthesis results in a 2 %c per 2-carbon acetyl group 
addition. These fractionations allowed the identifica- 
tion of fatty acids that were directly incorporated from 
symbiotic bacterial sources in mussels as opposed to 
those obtained through de novo synthesis (Murphy and 
Abrajano, 1994). 
In this study we compared the d 15 N, d L3 C, 6 34 S of bulk 
tissues, plus the 6 13 C of specific fatty acids among four 
guilds of fish plus anadromous Alosa spp. in a tidal 
freshwater stream on the East Coast of the United 
States. Our objective was to determine if anadromous 
fish, captured more than 40 km from the salt-wedge, 
were isotopically distinct from freshwater residents, and 
to determine if freshwater guilds showed the incorpora- 
tion of marine allochthonous organic material. 
Materials and methods 
Field collections by boat electrofisher were made in the 
tributaries and main-stem of the Rappahannock River, 
VA (within a 40 -mile area between Fredericksburg and 
Tappahannock, VA) during March and May 1997 and 
1998 (Fig. 1). The Rappahannock River is tidal in this 
region (tidal range: 0.1 to 1 meter) and shares many 
physicochemical characteristics with other tidal fresh- 
water rivers in the region (Garman and Nielsen, 1992). 
Fishes were collected and placed on ice in the field, 
transported back to the laboratory, and muscle tissue 
samples were taken, which were then dried for later 
analysis. Analysis of the sulfur and compound specific 
fatty acid samples took several years and were completed 
by 2002. 
The fishes were placed into four different guilds 
based on feeding strategies taken from Jenkins and 
Burkhead’s (1993) seminal work on Virginia freshwater 
fishes, plus an anadromous life cycle group (Table 1). 
Bulk isotope tissue analysis, elemental analyzer, 
and isotope ratio mass spectrometry 
Samples of dorsal muscle tissue were dried at 60°C for 
three days and homogenized in preparation for analy- 
sis. The tissues were then lipid extracted by refluxing 
them in dichloromethane for 35 minutes (Knoff et al., 
2002), except for those samples selected for compound 
specific analysis, which were soxlet extracted (see 
below; gas chromatography-mass spectrometry (GC- 
MS) and compound specific stable isotope analysis 
(CSIA)). One milligram (mg) of dried, lipid-extracted 
muscle was used for <5 13 C and <5 15 N analysis. Six mg was 
used for d 34 S analysis. A Carlo Erba elemental analyzer 
(EA) (Fisons/VG/Micromass, Manchester, UK) coupled 
to a Micromass Optima isotope ratio mass spectrom- 
eter (IRMS) (Fisons/VG/Micromass, Manchester, UK) 
was used to obtain 6 13 C, 6 15 N and d 34 S values. The <5 13 C 
and d 15 N were obtained concurrently, and <5 34 S was 
determined during separate analytical runs. 
