172 
Fishery Bulletin 107(2) 
than would be expected for a C3 dominated system. The 
anadromous Alosa spp. were also 13 C-enriched relative 
to other guilds. All of their 5 13 C values cluster between 
-22 %o and - 1 6 %c, whereas all other guilds range to ap- 
proximately -28%c range (the most 13 C-depleted values 
reflecting allochthonous production). This 13 C enrich- 
ment in Alosa spp. is not due to incorporating autoch- 
thonous freshwater production. The 13 C-enrichment is 
a signal from the marine environment from which the 
Alosa spp. biomass was derived. This interpretation 
is supported by the markedly 34 S-enriched values of 
the Alosa spp., which are in most cases l%c greater 
than any other fish in this study (6 34 S value of sulfur 
fixed from marine S0 4 in the ocean at present is highly 
enriched relative to freshwater [Kaplan et al., 1963]). 
Therefore, the 13 C enrichment of the Alosa spp. biomass 
(and other anadromous fishes) is due to a marine influ- 
ence, not an autochthonous influence. 
Of the guilds examined, predators show the highest 
d 34 S value after the Alosa spp., but are not significantly 
enriched in 13 C relative to other guilds. The elevated 34 S 
in predators (many of whom are piscivores) shows that 
more marine sulfur is incorporated by this guild rela- 
tive to others. The predator’s elevated 6 15 N values place 
them at the top of the fish food web, although some 
smaller individuals (blue catfish), feed at lower trophic 
levels while young (Jenkins and Burkhead, 1993). 
The link between anadromous Alosa spp. and the 
predators is also supported by the fatty acid carbon iso- 
tope signatures. Alosa spp. 16 and 18 carbon FAs were 
generally the most 13 C-enriched of the fish examined 
(Table 3). The two large (53cm TL) blue catfish show 
two very different FA isotope profiles. One blue catfish 
(B in Table 3) had a series of highly 13 C-enriched FAs 
(bulk muscle tissue d 13 C and 6 34 S are also enriched in 
this individual) and the other had FAs with isotope 
signatures similar to allochthonus primary production 
(also consistent with bulk muscle tissue <3 13 C and 5 34 S). 
Shorter chain (12 carbon) and more saturated FAs re- 
veal the original <5 13 C of the fats in the diet. Longer 
chain and unsaturated FAs can be subject to de novo 
transformations, which result in well established frac- 
tionations as chain length is systematically increased 
or as a double bond between carbons is created (making 
a point of unsaturation in a saturated FA). Generally, 
there is a 2 %c depletion in <5 13 C arising from each un- 
saturation and another 2 %c depletion for each two car- 
bon acetyl group addition (Deniro and Epstein, 1977). 
The most conservative tracer of dietary FAs, are the 
enriched precursors to long chain and unsaturated FAs. 
Among the FAs analyzed, the 12:0, 16:0, and 18:0 yield 
the best d 13 C estimate for dietary FAs, which clearly 
show distinct isotope signals depending on the carbon 
sources listed below: 1) 13 C-enriched marine isotope sig- 
nals (represented by alewife and blue catfish B), 2) al- 
lochthonus production (represented by blue catfish A), or 
3) a mix of autochthonous and allochthonus production, 
with the possibility of marine influences (represented by 
gizzard shad, although their 6 34 S values do not reflect 
the typical marine signal). 
The 6 13 C and 6 34 S distribution and range among the 
freshwater fishes suggest, not surprisingly, that both 
autochthonous and allochthonous nutrient sources, with 
the allochthonous sources being terrestrial C3 vegeta- 
tion and marine primary production inwelling to this 
tidal freshwater stream, more than 40 km from the 
Chesapeake Bay. Unlike streams on the West Coast 
of the United States, where marine derived nitrogen 
and carbon can be an important nutrient source to in- 
land ecosystems (Kline et al., 1993; Bilby et al., 2003; 
Chaloner et al., 2002), for all fish guilds in the study re- 
ported here, except the predators, there was not signifi- 
cant marine nutrient uptake. Several West Coast studies 
have shown that marine derived nitrogen, and some ma- 
rine derived carbon, contributed to invertebrates (Fran- 
cis et al., 2006; Hicks et al., 2005), primary producers, 
and juvenile fish within or near the sites receiving the 
spawning anadromous fish (Bilby et al., 2003; Koyama 
et al., 2005). For example, Bilby et al. (1996) found that 
17% and 30% of the nitrogen in collector-gathers and 
juvenile coho salmon (Oncorhynchus kisutch ) in Wash- 
ington, were derived from spawning salmon. Ben-David 
et al. (1998) found that salmon carcasses may have 
contributed to the nitrogen incorporated by some terres- 
trial plants, as well as deer mice, squirrels, and voles; 
and Wipfli et al. (2003) found that salmon carcasses 
fueled increased growth rates among young salmonids. 
However, those studies show that only some material 
from decaying salmon makes its way into invertebrates 
and riparian vegetation (Bilby et al., 1996, 1998; Fran- 
cis et al., 2006). There is strong evidence however, that 
the nutrients deposited as a result of the postspawn- 
ing death of anadromous adults did significantly sus- 
tain fry the following year (Bilby et al., 1996, 1998). i 
In the East Coast stream examined here, carnivores 
and generalists, which consume benthic invertebrates 
as part of their diet, did not show a marine signal. 
Compared with anadromous salmonids on the West 
Coast, East Coast herring have a lower postspawning 
mortality and their runs have less biomass. Both of 
these facts indicate that a limited amount of marine 
protein and nitrogen maybe be delivered to spawning 
streams unless it is consumed directly by predatory 
fish. This is consistent with findings suggesting ben- 
thic insects in Alosa spp. spawning streams do not 
accumulate large amounts of marine derived material, 
even if they are living closely with post-spawning anad- 
romous fish carcasses (Francis et al., 2006; Garman, 
1992). It should be noted that in West Coast streams 
associated with spawning salmon, invertebrate uptake 
can be substantial (Hicks et al., 2005; Chaloner et al., 
2002). Unlike most West Coast streams however, some 
tidal streams in Virginia have large piscivorous fish 
(introduced from Texas, Louisiana, or Mississippi in 
the 1970s) and these fish clearly incorporate marine 
material. So, while salmon (and presumably herring) on 
the West Coast import nutrients to the base of the food 
web (terrestrial autotrophs, young-of-the-year fish, and 
some invertebrates), in the steams examined here the 
marine material enters the top of the aquatic food web 
