546 
Fishery Bulletin 99(4) 
tive winter mortality has been documented among young- 
of-the-year (YOY) striped bass in both the Hudson River 
and Miramichi River populations (Hurst and Conover, 
1998; Bradford and Chaput 1 ). 
Young-of-the-year striped bass in the Hudson River expe- 
rience a winter energy deficit that varies in severity among 
years (Hurst et al., 2000). This interannual variability may 
be related to feeding conditions in the environment. Al- 
though studies of the summer feeding habits of YOY striped 
bass are available from most east coast estuaries (Markle 
and Grant, 1970; Boynton et ah, 1981; Gardinier and Hoff, 
1982; Rulifson and McKenna, 1987), information for over- 
wintering fish is extremely limited (Hartman and Brandt, 
1995a). A bioenergetics model has been developed for juve- 
nile striped bass (Hartman and Brandt, 1995b), describing 
metabolic rates and maximum consumption rates of fish 
across a wide range of temperatures. This type of determin- 
istic bioenergetic modeling does not take into account the 
observed variability in energy storage and depletion cycles 
(Hurst et ah, 2000) or compensatory feeding responses ob- 
served in response to depletion of reserves (Metcalfe and 
Thorpe, 1992). A better understanding of the energetics of 
overwintering juvenile striped bass, including in situ esti- 
mates of consumption rates, is required to fully evaluate 
the potential for winter starvation in this species. 
In our study, we present results of experiments measur- 
ing gastric evacuation rates at winter temperatures neces- 
sary to determine consumption rates of wild fish. We also 
describe the diet of overwintering YOY striped bass and 
estimate consumption rates of overwintering fish on 29 
dates over five winters. Finally, we analyze feeding pat- 
terns at both the individual and population level to deter- 
mine the factors regulating consumption rates of overwin- 
tering fish and discuss these results as they relate to the 
potential for winter starvation. 
Methods 
Gastric evacuation experiments 
Wild fish were captured from the Hudson River estuary 
and transported in river water to the Flax Pond Marine 
Laboratory of the State University of New York at Stony 
Brook in Old Field, New York. Fish were treated with 
0.60 ppm copper sulphate for 5 minutes and 15 ppm oxy- 
tetracycline for 5 days to reduce risk of mortality from 
infection. Fish were acclimated to laboratory conditions 
for at least 3 weeks prior to use and only fish appearing 
healthy and behaving normally were used in the experi- 
ment. Temperatures during the acclimation period were 
maintained between 1° and 5°C, and salinities were main- 
tained at 15 ppt. Fish were fed frozen adult brine shrimp 
1 Bradford, R. G.,and G. Chaput. 1997. Status of striped bass 
(Morone saxatilis) in the Gulf of St. Lawrence in 1996 and revised 
estimates of spawner abundance for 1994 and 1995. Dep. Fish, 
and Oceans, Can. Stock Assess Secretariat Res. Doc. 97/16, 31 p. 
Dep. Fisheries and Oceans, Science Branch, Gulf Region, RO. 
Box 5030, Moncton, New Brunswick E1C 9B6, Canada. 
(Artemia sp. ) and sand shrimp ( Crangon septemspinosa ) 
daily during the acclimation period. Prior to experimen- 
tation, groups of fish were acclimated to the test tem- 
perature (±0.5°C) for at least one week. Fish rarely fed 
voluntarily at low temperatures and were not offered food 
for between 3 and 5 days prior to use in the experiment. 
Evacuation rates were measured at 2°, 5°, 8° and 11°C. 
Temperatures in the experimental tanks were maintained 
by recirculating fluid chillers; they never differed from the 
prescribed temperature by more than 0.5° and were gener- 
ally within 0.2°C. Fish used in the experiment ranged in 
size from 88 to 150 mm TL (5.6 to 31.4 g wet weight), the 
natural size range of YOY striped bass in winter. To re- 
duce stress during the feeding process, fish were weighed 
(to 0.01 g wet weight) and acclimated to 65-L test tanks 
for 12 hours prior to feeding. Individual fish were captured 
with a dip-net from the test tank, force-fed the meal, and 
returned to the test tank within 1 minute. Fish were fed 
by opening the mouth and forcing the meal through the 
esophagus with a pair of blunt forceps. A meal comprised 
a single whole or partial C. septemspinosa weighing ap- 
proximately 2% of the fish’s body weight. A ration level of 
2% body weight was chosen because it falls in the upper 
range of, but is well below the maximum, gut fullness lev- 
els observed among wild fish. Each fish remained in the 
test tank until the time of sampling, when it was netted 
and sacrificed with an overdose of MS-222 anesthetic and 
measured (to 1.0 mm TL). The remaining stomach con- 
tents were dissected from the fish, dried of excess water, 
weighed (to 0.001 g), and dried to a constant weight at 
60°C. In no cases did fish regurgitate the meal following 
feeding or during the netting and sacrifice procedure. 
The evacuation rate of at least 3 fish was measured at 
each of 10 or more time points at each temperature (min. 
33 fish at 11°C; max. 43 fish at 2°C). The maximum inter- 
val between feeding and sacrifice encompassed the major- 
ity of the evacuation time at each temperature and ranged 
from 72 hours at 11°C to 168 hours at 2°C. The minimum 
interval used was 0.25 hours at all temperatures. Fish 
were exposed to a 10: 14 light:dark cycle to mimic the natu- 
ral winter photoperiod. Feeding time was standardized to 
the light:dark cycle, and fish were exposed to light for the 
first 8-10 hours of digestion. 
Evacuation of the meal was described with an expo- 
nential model allowing a time-lag prior to the beginning 
of evacuation because of its utility in estimating rations 
among wild fish (Elliott and Persson, 1978; Bromley, 1994). 
The model was fitted by using biphasic nonlinear regres- 
sion of untransformed data on the percentage of initial 
meal remaining over time with the equation 
100 if t < (c 0 + c{T) 
100e' l6 ° e6ir)[i_(Co+C|r)] if t>(c 0 + cfT), 
where T = temperature; 
t = time in hours since ingestion; 
b 0 and 6 X = the coefficients of the exponential relation- 
ship between evacuation rate and tempera- 
ture; and 
