386 
In vitro digestibility of some prey 
species of dolphins 
Kesko Sekiguchi* 
Peter B. Best 
Mammal Research Institute 
University of Pretoria, Pretoria 0002, South Africa 
Mailing address: Division of Natural Science, International Christian University 
3-10-2 Osawa, Mitaka-City, Tokyo 181, Japan 
*E-mail address: keikos@icu.ac.jp 
Studies of dolphin (Cetacea, Odon- 
toceti) food habits are conducted by 
examining stomach contents be- 
cause it is difficult to observe feed- 
ing behavior directly. It is rare, 
however, to find prey items intact 
in stomachs; often only fragments 
of muscle and some hard parts re- 
main. Identification of prey species 
and estimation of their original size 
are usually carried out with trace 
remains, such as cephalopod beaks 
(Clarke, 1980) and fish otoliths 
(Fitch and Brownell, 1968), because 
of their species-specific shapes and 
allometric relationships with body 
size (Clarke, 1962; Jobling and 
Breiby, 1986). 
There are several problems with 
using cephalopod beaks and fish 
otoliths in dietary studies. Otoliths 
are composed of calcium carbonate 
and can be eroded by stomach ac- 
ids (McMahon and Tash, 1979; da 
Silva and Neilson, 1985; Murie and 
Lavigne, 1985, 1986; Jobling and 
Breiby, 1986; Harvey, 1989). Reduc- 
tion in otolith size depends on the 
length of time they are exposed to 
stomach acids. Because otoliths are 
located inside the skull, the length 
of time they are exposed to acids 
may differ depending on the over- 
all digestibility of the fish species 
concerned. Some species are iden- 
tifiable even after their otoliths 
have been eroded and reduced in 
size. For such species, it may be 
difficult to tell if the otolith is of a 
reduced or original size (McMahon 
and Tash, 1979). Because estima- 
tion of fish prey size is usually 
based on a regression between 
otolith size and the weight or length 
of the prey, any reduction in otolith 
size that is not detected may cause 
prey size to be underestimated. 
The use of cephalopod beaks may 
create different problems. Although 
Bigg and Fawcett (1985) reported 
that soft-bodied squids ( Loligo 
opalescens ) decreased in weight 
faster than herring ( Clupea haren- 
gus pallasi) in an artificial diges- 
tion solution, cephalopod beaks 
were not dissolved by gastric acids. 
Cephalopod beaks may, therefore, 
accumulate in cetacean stomachs. 
It has been observed that some 
marine mammals occasionally re- 
gurgitate squid beaks (Clarke, 
1980; Pitcher, 1980). Cephalopod 
beaks present in a stomach may, 
consequently, represent the re- 
mains of more than one meal and 
thus may result in overestimations 
of the proportion of squid to fish in 
the predator’s diet. 
Bigg and Perez ( 1985) introduced 
the “modified volume” method to 
avoid the problem of the accumu- 
lation of cephalopod beaks. This 
method uses the frequency of occur- 
rence of nontrace remains to calcu- 
late the ratio between cephalopods 
and fish in a meal. However, if all 
prey remnants come from the same 
meal, any difference in digestibil- 
ity between prey items will affect 
the relative frequency of occurrence 
of nontrace remains when the stom- 
ach is examined. As an extreme 
case, prey items that are digested 
very rapidly would not be repre- 
sented by “nontrace remains” in the 
stomach soon after feeding. 
Differentials in digestion rates 
between Loligo squid and herring 
in an artificial digestion solution, 
as demonstrated by Bigg and 
Fawcett ( 1985), may apply to other 
prey species. For example, Jackson 
et al. (1987) could not detect differ- 
ences in the rates that fish and 
squid were completely digested in 
vitro but noted that exoskeletons of 
intact crustaceans resisted diges- 
tion. Thus, it is possible that diges- 
tion rates for each prey species, or 
prey type, could be used as “correc- 
tion factors” in dietary analysis. 
The present study investigates 
the differences in digestion rates of 
major prey species of dolphins in 
artificial digestion solutions. In 
addition, digestion rates of differ- 
ent sizes of the same prey species 
are considered. Digestion rates are 
then calculated to establish the 
basis for a revised method of di- 
etary analysis. 
Materials and methods 
The following fish and squid spe- 
cies were used in a set of six experi- 
ments: 1) 5 lanternfishes (Mycto- 
phidae), 5 large and 5 small Cape 
anchovies ( Engraulis capensis, 
Engraulidae); 2) 5 large and 5 small 
round herrings ( Etrumeus white- 
headi, Clupeidae); 3) 5 large and 5 
small pilchards ( Sardinops sagax, 
Clupeidae); 4) 5 hakes (Merluccius 
sp., Merlucciidae) and 5 chokka 
squids (Loligo vulgaris reynaudii, 
Loliginidae); 5) 5 maasbankers 
(horse mackerel) ( Trachurus tra- 
ehurus capensis, Carangidae) and 
Manuscript accepted 1 November 1996. 
Fishery Bulletin 95:386-393 (1997). 
