388 
Fishery Bulletin 95(2), 1 997 
addition of 10% HC1 (45 to 655 mL per 
experiment in total). The water bath 
rocked the container about 40 times per 
minute. As in the first experiment, each 
sample item was placed in a small mesh bag 
and suspended in the digestion solution. 
Every hour, each bag was lifted from the 
container, all excess liquid was wiped off 
with a paper towel, and the bag with 
sample weighed to the nearest 0.1 g. The 
physical appearance of each sample was 
also recorded. Weighings were made at 1- 
h intervals until the sample mass (i.e. 
measured weight minus weight of the 
empty mesh bag) had decreased to 5-10% 
of its original mass. 
To compare digestion rates between each 
species, the mean time to reach 20% of 
original weight (T., 0 ) was calculated for 
each sample species. This percentage was 
chosen because the rate of decline in mass 
decreased when the sample reached this 
point. This decrease probably resulted 
from inaccuracies in weighing smaller 
masses as well as from the accumulation 
of less digestible remains. The T., 0 values 
for different size groups of the same prey 
species were compared first with a Ltest. 
Then, one-way ANOVA and the Newman- 
Keuls test were applied to compare all 
sample species (Zar, 1974). A digestion rate 
ratio was calculated from the T 20 values 
for each species, expressed as a proportion 
of that for lanternfish. 
Results 
100 
100 
c 
to 
E 
CD 
cc 
100 
Time (h) 
Figure 1 
The rate of digestion of fish and squid species in an artificial digestion 
solution, expressed as the percentage of their original weight remain- 
ing at hourly intervals. The slope shown is the digestion rate to the 
mean time to 20% of the initial weights (T 20 ). (A) 10 lanternfish 
Samples were digested almost completely 
in the pepsin solution. Although digestion 
rates were quite different among species, 
the sequences of digestion of particular 
tissues were similar among species (Table 
2). Although the head of a fish usually dis- 
integrated when about half the body had been di- 
gested, otoliths were not always visible through the 
mesh bag at this stage. In the case of hake and 
maasbanker, the dorsal surface of the head began to 
be digested at an earlier stage (15% digested at 2-3 
h for hake, 5-6 h for maasbanker) than that found 
for other fish species. Otoliths became visible 
(through the mesh bag) at 5-8 h for hake and at 19 h 
for maasbanker. Hake otoliths fell through the mesh 
at 9-13 h. Most otoliths were dissolved completely 
when the experiments with hake terminated at 20 h 
(Myctophidae sp.), (B) 5 maasbanker (Trachurus t. capensis), (C) 5 hake 
(Merluccius sp.), (D) 5 pelagic goby (Sufflogobius bibarbatus), (E) 5 
chokka squids ( Loligo v. reynaudii), and (Fl 5 red squids ( Todaropsis 
eblanae). 
(except one otolith), and at 27 h (except for five 
otoliths) with maasbanker. Some otoliths of reduced 
size were recovered in experiments involving other 
species (i.e. 16 from anchovy, 8 from herring, and 10 
from goby). All squid beaks recovered at the termi- 
nation of the experiments showed no obvious signs of 
having been digested. 
All samples decreased in weight over time (h), each 
species having different rates of digestion (Figs. 1 
and 2). Lanternfish were digested very quickly, and 
were almost completely gone within 9 hours. Hake 
