Butler et al. : Feeding ecology of Thunnus thynnus in North Carolina 
65 
to our estimate. The only other GER for a tuna spe- 
cies that we are aware of is that of Olson and Boggs 
(1986). Those authors found that under laboratory 
conditions, the GER of yellowfin tuna was best repre- 
sented by a linear evacuation model and depended on 
the type, surface area, and digestibility of the prey 
consumed. Their results showed that mackerel, which 
contained the highest lipid level of the prey types 
they examined, were the most digestion-resistant and 
consequently took the longest (~18.5 hours) to evacu- 
ate. Because the diets of bluefin tuna in our study 
were dominated by Atlantic menhaden by weight, our 
GER estimates are representative of Atlantic menha- 
den prey and should be used with caution if applied 
to bluefin tuna that feed on other prey, particularly 
prey that may differ in digestibility (e.g., prey with 
an exoskeleton, and having differing lipid levels). 
In the present study, several key assumptions 
were made to estimate GER. Because no fish were 
landed from 1900 to 0300 EST, we assumed that 
feeding was negligible at night. If bluefin tuna do 
feed throughout the night, then our GER and mean 
gut fullness values (estimated over a 24-hour period) 
could potentially be biased low (i.e., peak gut fullness 
would occur later in night with a shorter [and faster] 
time to digest prey). We also had to assume that 
stomachs were not completely emptied several hours 
before the first presunrise samples were collected. 
Given that many of the samples collected during 
presunrise hours contained prey in the final stages 
of digestion, this appears to be a valid assumption. 
If digestion was completed before the time of sunrise 
collections (shortening the time between peak and 
valley), then our current estimates of GER would be 
biased low (i.e., we would again be assuming a longer 
time to digest prey than what is actually true). 
Estimates of digestion times in pen-held and wild 
Atlantic bluefin tuna corroborate our estimates of 
GER. Butler and Mason (1978) used stomach content 
analysis on pen-held giant bluefin tuna (>200 kg) 
that were fed a variety of “forage fish” species and 
determined that it took 18-20 hours to completely 
empty a full stomach (other than viscous liquid). 
Using acoustic telemetry, Stevens et al. (1978) fed 
mackerel to captive giant bluefin tuna (>200 kg) 
and identified gradual increases of stomach tempera- 
tures which lasted 14 to 20 hours following a feeding 
event. Similarly, Carey et al. (1984) used acoustic 
transmitters in pen-held fish to measure stomach 
contractions and temperature increases after bluefin 
tuna feeding events (prey=mackerel and herring) 
and concluded it took 18 to 21 hours before liquefied 
food was moved out of the stomach and into the in- 
testine and pyloric caeca These studies indicate that 
complete evacuation time is not reached in less than 
18 to 20 hours and support our estimated digestion 
time of ~20 hours (i.e., from peak gut fullness to gut 
fullness near zero). 
The DR of giant bluefin tuna has been estimated in 
several studies by various methods. Palomares and Pauly 
Figure 3 
Diel patterns of mean gut fullness (kg prey/kg predator • 
100%; ± standard error) for (A) large medium, (B) giant, 
and (C) pooled bluefin tuna ( Thunnus thynnus) collected 
off Cape Lookout, North Carolina, during the winters of 
2004-05 and 2005-06. Open circles represent values used 
to estimate gastric evacuation; filled triangles represent 
values that were not used during gastric evacuation esti- 
mates (see Methods section for description of data selec- 
tion). The fit of the exponential gastric evacuation model 
(S t =S 0 -e~ GER f ) is represented by the solid line, where S t 
is the individual stomach fullness (kg prey/kg predator) 
at time t; S 0 is the stomach fullness at time f=0; GER is 
the instantaneous rate of gastric evacuation (rate per hr); 
and t is the time in hours after peak gut fullness. Samples 
sizes are listed above the mean gut fullness value. Time 
point 0 = 1200-1300 hours for (A) and 0 = 1300-1400 hours 
for (B); average time between sunset and sunrise is rep- 
resented by solid horizontal bar. 
(1989) estimated a DR of 1.08% body mass per day for 
maximum-size (e.g., L CO =332.0 cm) bluefin tuna, using 
