Journal of Mammalogy, 89(5):1248–1256, 2008
FOOD HABITS OF FINLESS PORPOISES NEOPHOCAENA
PHOCAENOIDES IN WESTERN KYUSHU, JAPAN
MIKI SHIRAKIHARA,* KENJI SEKI, AKIRA TAKEMURA, KUNIO SHIRAKIHARA, HIDEYOSHI YOSHIDA,
AND
TAKESHI YAMAZAKI
Faculty of Science, Toho University, Miyama, Funabashi, Chiba 274-8510, Japan (MS)
Kyusyu Kaihatsu Engineering Co., Ltd., Nishikigaoka, Kumamoto 862-0912, Japan (KS)
Department of Fisheries, Nagasaki University, Bunkyo-machi, Nagasaki 852-8521, Japan (AT)
Ocean Research Institute, The University of Tokyo, Minamidai, Nakano-ku, Tokyo 164-8639, Japan (KS)
National Research Institute of Far Seas Fisheries, Fukuura, Kanazawa-ku, Yokohama 236-8648, Japan (HY)
Laboratory of Animal Management and Resources, Graduate School of Bioagricultural Science,
Nagoya University, Nagoya, Aichi 464-8601, Japan (TY)
We examined the stomach contents of 87 finless porpoises (Neophocaena phocaenoides) in the Ariake Sound–
Tachibana Bay area and Omura Bay in western Kyushu, Japan, between 1987 and 1992. Fish (Gobiidae and
Atherinidae) were the most numerous and most frequently occurring prey in Omura Bay, whereas both
cephalopods (Octopodidae, Sepiidae, Sepiolidae/Sepiidae, and Loliginidae) and fishes (Clupeidae, Engraulidae,
and Sciaenidae) were equally important in Ariake Sound–Tachibana Bay. Species compositions in the
commercial catch differed between the 2 waters, suggesting that differences in prey availability may explain the
geographical variation in diet. Finless porpoises in Ariake Sound–Tachibana Bay showed ontogenetic and
seasonal variations in diet. The mean length at weaning was estimated to be 101 cm, corresponding to
approximately 6 months of age. Calves fed on small-sized demersal fish and cephalopods. The predominant prey
species for sexually mature individuals (including lactating females) consisted of konoshiro gizzard shad
(Konosirus punctatus) and cephalopods. Seasonal availability of these organisms may be related to births in the
fall–winter season in Ariake Sound–Tachibana Bay. Size-dependent prey selectivity was equivocal. Day–night
difference in foraging time was not indicated by the index of stomach fullness.
Key words:
bycatch, calving season, cetaceans, diet, finless porpoise, Japan, Neophocaena phocaenoides
Finless porpoises (Neophocaena phocaenoides) are distributed in coastal or shallow waters and some rivers in the IndoPacific region (Amano 2002). It is known that finless porpoises
prey on fish, cephalopods, and crustaceans (Barros et al. 2002;
Chen et al. 1979; Park et al. 2005; Pilleri and Gihr 1972; Zhou
et al. 1993), but geographical variation exists in their diet. In
Chinese waters, Yangtze finless porpoises fed on bottom fish
or middle- and lower-layer fish, such as Cyprinus carpio,
Xenocypris davidi, and Pseudobagrus (Chen et al. 1979), and
coastal finless porpoises preyed on icefish (Salanx), mullets
(Liza), common sea bass (Lateolabrax japonicus), common
mackerel (Pneumatophorus japonicus), prawns (Penaeus), and
squid (Loligo—Zhou et al. 1993). In Hong Kong waters, 25
species of fish, 3 genera of cephalopods, and a shrimp were
detected in stomachs of finless porpoises (Barros et al. 2002).
* Correspondent: [email protected]
Ó 2008 American Society of Mammalogists
www.mammalogy.org
1248
In contrast, in Korean waters, 13 species of crustaceans,
8 species of fish, and 3 species of cephalopods were identified
as prey organisms (Park et al. 2005). In Japanese waters,
stomach contents of finless porpoises were examined using
a small number of specimens in a limited number of waters
(Kataoka et al. 1976; Masaki 1980; Shirakihara et al. 1992b).
No quantitative analyses of food habits have been published.
The distribution of porpoises within Japanese waters is
discontinuous (Shirakihara et al. 1992a), and multiple populations have been genetically identified (Yoshida et al. 2001).
Differences in calving season among populations also have
been reported (Shirakihara et al. 1993). In western Kyushu,
there are 2 populations: the Ariake Sound–Tachibana Bay population and the Omura Bay population. Ariake Sound is a semiclosed inlet with a mean depth of about 20 m (Fig. 1). The
innermost area has many rivers and a vast tidal flat, whereas
the area at the mouth is deep (the maximum depth 160 m)
and is connected to Tachibana Bay through Hayasaki Strait.
Tachibana Bay, which has a flat bottom over most of its extent,
is characterized as a transition area between Ariake Sound and
the open sea. In contrast, Omura Bay is an enclosed coastal sea
October 2008
SHIRAKIHARA ET AL.—FOOD HABITS OF FINLESS PORPOISES
1249
FIG. 1.—a) Map of the study area. The coast where the specimens of finless porpoises (Neophocaena phocaenoides) were collected is indicated
with a bold line. b) Main habitat of the finless porpoises in Japan (after Shirakihara et al. 1992b).
with a mean depth of 15 m. Prey availability may differ among
the 3 bodies of water.
In our study, we described the stomach contents of finless
porpoises in 2 populations in western Kyushu. Then we discuss
geographical variation in the diet with attention to the calving
season.
MATERIALS AND METHODS
Specimens for this study were collected on the coast of
Ariake Sound–Tachibana Bay area and Omura Bay between
1987 and 1992 (Fig. 1). We distributed leaflets among the fisheries cooperative associations near the fishing ports and asked
fishermen to bring us the carcasses of bycaught finless porpoises
for biological examination. When a specimen was provided,
a questionnaire survey was carried out on the fishing method:
soak time of the net, month at which the net was used, fish
species caught, and so on. We also collected stomach contents
from animals stranded or found drifting. We examined stomach
contents of 78 porpoises (67 bycaught, 5 stranded, and 6 found
drifting) from the Ariake Sound–Tachibana Bay area and 9
bycaught animals from Omura Bay. We met guidelines
approved by the American Society of Mammalogists (Gannon
et al. 2007).
Porpoise carcasses were transported to the laboratory, where
external measurements were taken and dissections took place.
The entire stomach was removed from each carcass and
weighed full and empty to the nearest 1 g. The contents of the
fore and main stomachs were recovered by rinsing them in
a sieve with a mesh size of 0.8 mm. The presence of milk was
confirmed visually. When the intact or partially digested body
of a fish was found, its standard length was measured in
millimeters. In addition to external morphology, hard parts—the
vertebral column, otoliths, cuttlebones, and cephalopod
beaks—also were used in prey identification, which was done
with the aid of published guides (Clarke 1986; Masuda et al.
1984; Matsubara 1979; Ohe 1985; Takahashi 1962) and
a reference collection of otoliths assembled from fishes obtained
at local fish markets. Information on the reference collection is
available from MS. Identification of otoliths that were not included the collection was supported by Fumio Ohe and his collection including 1,200–1,300 species of fish (Ohe 1985). For
cephalopod beaks of Octopodidae, Sepiidae, Sepiolidae, and
Loliginidae found in stomachs of porpoises (see ‘‘Results’’),
lack of a complete reference collection of all species at different
ontogenetic stages, similarity in beak morphology, or small size
of beaks prevented identification to species level. Small-sized
beaks of Sepiidae and Sepiolidae were pooled.
The number of prey consumed by each porpoise was calculated by adding the number of intact or partially digested prey
to the number estimated from identified remains found in
the stomach (the number of sagittal otoliths divided by 2 for
fish and the greatest number of upper or lower beaks for
cephalopods).
The relative importance of each prey type was determined by
the number of individuals of a particular prey type divided by
the number of individuals of all prey (numerical abundance
[%N]) and by the number of porpoises that had eaten each prey
type divided by the number of porpoises that had eaten solid
food (frequency of occurrence [%F]).
The relative importance of each prey type in terms of mass
was calculated for 5 species of fish that were dominant
according to the %N or %F data: Japanese pilchard (Sardinops
melanostictus), Japanese sardinella (Sardinella zunasi), ko-
1250
TABLE 1.—Relationship between otolith length (OL in mm) and
standard length (SL in mm) or weight (WT in g) of 5 species of fish
obtained from local fish markets.a
Species
Regression
R
n
Sardinops melanostictus
SL ¼ 32.49 þ 57.23OL
lnWT ¼ 0.38 þ 4.03(ln OL)
SL ¼ 1.25 þ 34.88OL
lnWT ¼ 0.22 þ 3.15(ln OL)
SL ¼ 18.38 þ 46.91OL
lnWT ¼ 3.12(ln OL)
SL ¼ 18.99 þ 35.34OL
lnWT ¼ 1.23 þ 4.15(ln OL)
SL ¼ 36.07 þ 23.45OL
lnWT ¼ 1.80 þ 4.07(ln OL)
0.90
0.89
0.96
0.97
0.96
0.96
0.97
0.95
0.97
0.99
67
67
47
47
52
52
37
37
40
42
Sardinella zunasi
Konosirus punctatus
Engraulis japonicus
Pennahia argentata
TABLE 2.—Number of finless porpoises (Neophocaena phocaenoides) taking solid food in western Kyusyu, Japan, based on stomach
contents of porpoises collected between 1987 and 1992.
Population
Ariake SoundTachibana Bay
Growth stage
Calves
Juveniles
Lactating females
Other sexually mature
individuals
Total
a
a
Vol. 89, No. 5
JOURNAL OF MAMMALOGY
R ¼ correlation coefficient; n ¼ sample size.
b
c
noshiro gizzard shad (Konosirus punctatus), Japanese anchovy
(Engraulis japonicus), and white croaker (Pennahia argentata). We measured the maximum length of undigested otoliths
(from the tip of the rostrum to the posterior margin) of the 5
species of fish and estimated body weight using the equations
listed in Table 1 (based on samples purchased in fish markets).
The mass of each fish found in a stomach was calculated as the
product of the mean mass of the restored fishes and the number
of individuals of the species consumed. If there were no
undigested otoliths in a stomach, the mean mass was estimated
from all the undigested otoliths in all the stomachs.
We classified porpoises as calves, juveniles, or sexually
mature individuals on the basis of age and reproductive condition, following Shirakihara et al. (1993). We defined a calf as an
individual , 1 year old, and a juvenile as an individual 1 year
old and sexually immature. We regarded the period from March
to August as the spring–summer season and that from
September to February as the fall–winter season, accounting
for water temperature in Ariake Sound, which is highest in
August (26–288C) and lowest in February (8–98C—Inoue
1980). Differences in mean number of prey items, prey taxa,
and wet weight between populations, ontogenetic stages
(juveniles and sexually mature individuals), and seasons were
tested with the nonparametric Wilcoxon rank sum test. Contingency table analysis was used to test the difference in the
composition of the 12 species with a high level of %N or %F.
To examine whether larger porpoises selectively feed on
larger prey, linear regression analysis was conducted using
body length of porpoises and standard length of fish. We
pooled data on the standard length of fish reconstructed from
the otolith length and undigested fish because the difference
between them was not significant (W ¼ 269, P ¼ 0.0769 for
konoshiro gizzard shad; W ¼ 259, P ¼ 0.0555 for Japanese
sardinella). To estimate when finless porpoises feed during
a day, we compared an index of stomach fullness (wet weight
of stomach content/body weight of the porpoise) of the
porpoises taken during 0600–1600 h (daytime) and during
1600–0600 h (night). To estimate the body length at which half
the individuals are expected to wean or begin taking solid food,
we used a standard logistic equation following Archer and
Robertson (2004):
a
b
Sp-Su
Omura Bay
Sp-Sua Fa-Wib Total
Fa-Wi
Total
8
5
0
12
16
3
20
21
3
0
2
1
3
12
15
4c
0
4
16
43
59
7
2
9
1
1
0
1
3
1
Spring–summer season (March–August).
Fall–winter season (September–February).
Including female estimated mature based on age of 14 years.
P¼
1
;
1 þ eabx
where P is the probability that milk or food is present in
a stomach, x is length, and a and b are parameters. For each
specimen, P was recorded as 0 if milk or food was absent in
a stomach or 1 if present. Parameters were estimated by
a maximum-likelihood procedure. This analysis and other
statistical analyses were made with the software JMP 6.0.3
(SAS Institute Inc., Cary, North Carolina).
RESULTS
Diet of porpoises in the Ariake Sound–Tachibana Bay
population.—Among the 78 porpoise stomachs examined,
2 were completely empty. Only milk was found in 17
stomachs. Solid food was found in 59 stomachs, of which 3
also contained milk. Stomachs of calves and lactating females
accounted for one-third and 5% of the sample, respectively
(Table 2). The number of stomachs collected in the fall–winter
season was more than twice the number in the spring–summer
season, although the fishing nets were used every month (v2 ¼
19.413, d.f. ¼ 11, P ¼ 0.0541). This may be related to seasonal
variation in density of the porpoises, which showed a minimum in August in the middle of Ariake Sound (Shirakihara
et al. 1994).
A total of 3,267 individual prey items was found, representing 20 species of fish, 3 species of cephalopods, and 1 species of
crustacean (Table 3). Cephalopods accounted for 66.0% of the
total number of prey consumed and occurred in 54 (91.5%) of
59 stomachs examined. The following 5 species of fish showed
high levels of %N and %F: Clupeidae (Japanese pilchard,
Japanese sardinella, and konoshiro gizzard shad), Engraulidae
(Japanese anchovy), and Sciaenidae (white croaker). Small
demersal fish (Gobiidae and Apogonidae) accounted for .10%
of the total number of prey, with %F of 13.6% and 22.0%,
respectively. Among cephalopods, 4 prey taxa (Octopodidae,
Sepiidae, Sepiolidae/Sepiidae, and Loliginidae) were dominant,
with .12%N and 30%F. Shrimp made up 2.5%N and 23.7%F,
October 2008
SHIRAKIHARA ET AL.—FOOD HABITS OF FINLESS PORPOISES
1251
TABLE 3.—The numerical abundance (number of individuals of a particular prey type divided by the number of individuals of all prey [%N])
and frequency of occurrence (number of porpoises that had eaten a prey type divided by the number of porpoises that had eaten solid food [%F])
of prey items of finless porpoises in Ariake Sound–Tachibana Bay and Omura Bay based on stomach contents of porpoises collected between
1987 and 1992.
Ariake SoundTachibana Bay
All individuals
%N
(n ¼ 3,267)
Fish
Etrumeus teres
Sardinops melanostictus
Sardinella zunasi
Konosirus punctatus
Engraulis japonicus
Congridae
Ariosoma meeki
Gnathophis sp.
Muraenichthys sp.
Hyporhamphus sajori
Hypoatherina valenciennei
Chelon sp.
Chelon affinis
Sphyraena pinguis
Lateolabrax japonicus
Apogonidae
Trachurus japonicus
Leiognathus rivulatus
Pennahia argentata
Nibea albiflora
Cepolidae
Acanthuridae
Psenopsis anomala
Gobiidae
Parachaeturichthys polynema
Odontamblyopus lacepedii
Pinguipedidae
Parapercis multifasciata
Parapercis sexfasciata
Sebastes inermis
Callionymidae
Unidentified fish
Cephalopods
31.2
0.1
Total
91.5
2.5
0.2
0.1
100.0
21.2
0.6
0.1
0.9
2.8
10.0
5.0
10.0
10.0
0.1
5.0
5.0
Mature individuals
%F
(n ¼ 21)
%N
(n ¼ 642)
90.5
44.4
6.8
0.1
4.0
5.7
19.0
9.5
38.1
4.8
0.1
4.8
0.1
%F
(n ¼ 18)
100.0
0.2
1.7
14.5
12.9
5.1
0.5
0.2
Omura Bay
%N
(n ¼ 712)
90.6
5.6
11.1
11.1
44.4
22.2
5.6
5.6
4.8
0.1
5.3
35.0
0.6
10.0
0.1
5.0
25.9
0.1
0.5
15.0
5.0
5.0
1.4
58.8
%F
(n ¼ 9)
88.9
0.1
0.1
90.0
1.6
2.8
23.7
1.7
5.1
100.0
respectively. The above-mentioned prey items were regarded as
important prey species in %N or %F.
The konoshiro gizzard shad was clearly dominant in relative
importance by mass, accounting for 53.6% of the total
reconstructed mass of the 5 most important fish species
4.8
4.8
19.0
4.8
0.5
0.3
0.4
19.0
9.5
4.8
1.3
23.8
0.1
0.1
4.8
9.5
4.8
33.3
95.2
16.8
0.2
8.9
26.6
0.2
14.3
1.1
25.0
30.0
2.8
0.1
0.1
0.6
0.1
75.0
25.0
5.0
20.0
60.0
10.0
50.0
10.0
11.1
11.1
3.3
0.4
0.1
100.0
14.3
42.9
5.6
0.1
1.0
11.1
11.1
0.6
11.1
0.1
11.1
3.9
0.2
22.2
5.6
0.5
5.6
81.7
55.6
0.6
33.3
0.3
1.7
5.6
5.6
2.0
38.9
88.9
38.9
5.6
33.3
27.8
22.0
0.5
50.0
16.7
1.1
33.3
2.0
2.1
22.2
11.1
0.5
100.0
4.4
5.8
0.2
22.7
3.0
0.5
38.1
4.8
4.8
11.1
33.3
0.2
55.1
57.1
4.8
47.6
66.7
4.8
66.7
19.0
7.0
3.8
30.0
6.5
4.8
0.3
0.6
25.0
19.5
0.2
15.1
14.5
2.2
5.1
0.6
20.3
28.8
%N
(n ¼ 1,588)
0.3
40.7
5.1
33.9
52.5
5.1
55.9
15.3
4.1
2.8
Juveniles
%F
(n ¼ 20)
75.0
0.1
1.7
1.7
1.7
1.7
22.0
1.7
3.4
13.6
5.1
1.7
1.7
1.7
13.6
1.7
1.7
1.7
3.4
1.7
1.7
1.7
32.2
15.5
0.2
13.6
18.1
0.8
12.9
0.8
Shrimp
Oratosquilla oratoria
Isopoda
38.4
3.4
66.0
%N
(n ¼ 1,037)
1.7
13.6
8.5
30.5
11.9
1.7
1.7
1.7
1.7
0.0
0.0
0.0
0.0
0.0
2.1
0.0
0.2
1.0
0.2
0.2
0.0
0.1
8.8
0.0
0.2
0.0
0.1
0.1
0.3
0.2
1.1
Unidentified squid
Crustaceans
88.1
0.0
3.8
2.9
4.8
4.7
0.1
0.0
0.0
0.0
0.0
Octopodidae
Octopus minor
Sepiidae
Sepiolidae/Sepiidae
Sepiella japonica
Loliginidae
Todarodes pacificus
Calves
%F
(n ¼ 59)
22.2
22.2
0.3
5.1
11.1
22.2
33.3
4.8
33.3
0.3
11.1
100.0
(21,466 g). White croaker accounted for 27.7%, and the other
3 species together accounted for ,20% of the mass (12.1%
for Japanese pilchard, 4.4% for Japanese anchovy, and 2.3%
for Japanese sardinella). The mean number of prey items,
number of prey taxa, and wet weight were 55.4 (SD ¼ 66.8,
1252
Vol. 89, No. 5
JOURNAL OF MAMMALOGY
range ¼ 1–352), 4.5 (SD ¼ 2.9, range ¼ 1–11), and 188.0 g
(SD ¼ 203.4 g, range ¼ 1–809 g), respectively.
Ontogenetic variation in diet of porpoises in the Ariake
Sound–Tachibana Bay population.— The %N and %F of prey
items by ontogenetic stage are shown in Table 3. The %N of
cephalopods exceeded 50%, with a maximum of 75.0% for
juveniles. The %N of fish was the highest for sexually mature
individuals (44.4%). The %N of crustaceans was low for each
stage, and shrimp were not found in the stomachs of sexually
mature individuals. The %F of fish was 75.0% for calves,
90.5% for juveniles, and 100.0% for sexually mature
individuals. The %F of cephalopods reached approximately
90% in each stage.
For calves, small-sized demersal fish such as Gobiidae and
cephalopods such as Octopodidae and Sepiidae were ranked in
the top 3 for %N. Apogonidae (35.0%F), Sepiolidae/Sepiidae
(60.0%F), and Loliginidae (50.0%F) had high levels of %F.
For juveniles, cephalopods were the predominant prey items,
but %N and %F of Clupeidae were slightly above those for
calves. For sexually mature individuals, konoshiro gizzard
shad, Sepiidae, and Loliginidae had the high levels of %N or
%F. Approximately one-fifth of the sexually mature individuals
fed on white croaker. No Gobiidae were found in the stomachs
of sexually mature individuals. A significant difference was
detected in the composition of 12 selected important prey items
(7 fishes, 4 cephalopods, and 1 crustacean) among the
ontogenetic stages (v2 ¼ 1,561.56, d.f. ¼ 22, P , 0.0001).
There was no significant difference between juveniles and
sexually mature individuals in the mean number of prey taxa
(W ¼ 312.5, P ¼ 0.182) or in the mean wet weight of stomach
contents (W ¼ 274, P ¼ 0.7028), although the mean number of
prey items for juveniles was larger than that for sexually mature
individuals (W ¼ 287.5, P ¼ 0.0425).
The standard length of the fish consumed ranged from 5.3 to
32.5 cm. The larger porpoises tended to feed on smaller fish
(F ¼ 38.05, d.f. ¼ 1, P , 0.0001), but the same analysis using
the data for konoshiro gizzard shad ranging from 10.8 to 19.9
cm or Japanese anchovy ranging from 7.1 to 10.3 cm showed
that the larger porpoises tended to prey on larger shad (F ¼
6.90, d.f. ¼ 1, P ¼ 0.0136). No significant difference was
detected for Japanese anchovy (F ¼ 2.49, d.f. ¼ 1, P ¼
0.1182). Taking into consideration that a juvenile porpoise
with a body length of 134.0 cm consumed a 32.0-cm conger
(Gnathophis) and another calf, 108.0 cm long, fed on a 32.5-cm
green eelgoby (Odontamblyopus lacepedii), size-dependent
prey selectivity may be not clear for finless porpoises.
Onset of feeding in the Ariake Sound–Tachibana Bay
population.— Only milk was found in 17 stomachs of calves
, 99.5 cm in body length. We found Sepiolidae/Sepiidae and
milk in the stomachs of 3 individuals 99.5 cm, 100.5 cm, and
107.0 cm in body length. Milk was not detected in the stomach
of a calf 93.5 cm in body length, which had fed on Apogonidae
and Gobiidae. Two other calves, each 104.0 cm in body length,
fed on unidentified fish and Sepiolidae/Sepiidae. The calves
had preyed on Gobiidae, Apogonidae, or Sepiolidae/Sepiidae,
suggesting that they begin to feed by taking small-sized
individuals of demersal fishes, cuttlefishes, or bobtail squid.
FIG. 2.—The numerical abundance (%N) and frequency of
occurrence (%F) of prey items of finless porpoises (Neophocaena
phocaenoides) in the Ariake Sound–Tachibana Bay population by
season. Circular graph shows %N of fish (Fi), cephalopods, and
crustaceans (Cr).
The logistic regression analysis yielded the following
equations of the probability of milk or food:
PðmilkÞ ¼
1
1 þ e36:646þ0:362x
and
PðfoodÞ ¼
1
1þ
e48:4790:494x
;
where x is body length of the porpoise in centimeter. Body
length at which 50% of individuals wean or take solid food was
estimated to be 101 cm (95% confidence interval [95% CI] ¼
95–105 cm) and 98 cm (95% CI ¼ 92–103 cm).
Seasonal variation in diet of porpoises in the Ariake Sound–
Tachibana Bay population.—The %N of cephalopods was
76.5% in the spring–summer season and 60.4% in the fall–
winter season, respectively (Fig. 2). The %F of cephalopods
was 88.4% in the fall–winter season, whereas all the porpoises
fed on cephalopods in the spring–summer season. The %N
values of Clupeidae and Sepiidae were high in the fall–winter
season, whereas those of Octopodidae and Sepiolidae/Sepiidae
were high in the spring–summer season. The %F of konoshiro
gizzard shad and Sepiidae showed a high level in the fall–
winter season. A significant difference was detected in the
composition of 12 selected important prey items between the
seasons (v2 ¼ 565.05, d.f. ¼ 11, P , 0.0001). No significant
difference was detected between the 2 seasons in mean number
of prey items (W ¼ 511, P ¼ 0.6029), prey taxa (W ¼ 505.5,
P ¼ 0.6673), or wet weight (W ¼ 242, P ¼ 0.149).
Three lactating females were collected in the fall–winter
season. One who was simultaneously pregnant had consumed
20 individual shad. This was the highest number consumed by
a porpoise in the sample. Another lactating female had the
heaviest stomach contents (wet weight 809 g) in the sample and
preyed on 17 individual shad (mean estimated body length
17.8 cm, mean estimated body weight 91 g). Shad showed
the highest %N (27.4%), followed by Loliginidae (26.7%),
October 2008
SHIRAKIHARA ET AL.—FOOD HABITS OF FINLESS PORPOISES
1253
(W ¼ 47, P ¼ 0.3191), and wet weight (W ¼ 38.5, P ¼ 0.6678)
were detected between the 2 populations.
DISCUSSION
FIG. 3.—The numerical abundance (%N) and frequency of
occurrence (%F) of prey items of juveniles and sexually mature
individuals of finless porpoises (Neophocaena phocaenoides) in the
Ariake Sound–Tachibana Bay population and Omura Bay population
in spring–summer season. Circular graph shows %N of fish (Fi),
cephalopods, and crustaceans (Cr).
Octopodidae (18.5%), and Sepiolidae/Sepiidae (11.9%) in
lactating females.
Foraging time in the Ariake Sound–Tachibana Bay
population.— The mean of the index of stomach fullness was
not significantly different between day and night (W ¼ 51, P ¼
0.8016). The means for day and night were 0.43% (n ¼ 6,
SD ¼ 0.31%, range ¼ 0.12–0.90%) and 0.63% (n ¼ 11, SD ¼
0.60%, range ¼ 0.02–1.92%), respectively. The maximum
value was obtained at night. Xiao et al. (2005) mentioned that
no clear diurnal rhythm was found in behavior of finless
porpoises in captivity.
Diet of porpoises in the Omura Bay population.— All of the
stomachs of 9 porpoises (1 calf, 3 juveniles, and 5 sexually
mature individuals) contained food remains. Fish made up
90.6%N and 88.9%F (Table 3). Gobiidae with 81.7%N and
55.6%F and Atherinidae (Sumatran silverside [Hypoatherina
valenciennei]) with 6.5%N and 33.3%F showed the major
contributions, but Clupeidae and cephalopods were present at
low levels.
Differences in diet between populations.— Because ontogenetic and seasonal variations were found in the diets of
porpoises in the Ariake Sound–Tachibana Bay area, we compared the %N of fish, cephalopods, and crustaceans of juveniles
and sexually mature individuals collected in the spring–summer
season between the populations (Fig. 3). Cephalopods had
higher %N in the Ariake Sound–Tachibana Bay area (81.0%),
whereas fishes were prominent in Omura Bay (90.5%). The next
highest-level taxon was fish in the former area and crustaceans
in Omura Bay. Clupeidae (except konoshiro gizzard shad),
white croaker, Octopodidae, Sepiidae, and Sepiolidae/Sepiidae
were absent in stomachs of finless porpoises in Omura Bay. The
composition of 12 important prey items was significantly
different between the 2 populations (v2 ¼ 1,342.6, d.f. ¼ 11,
P , 0.0001). No significant differences in the 3 parameters of
mean number of prey items (W ¼ 54.5, P ¼ 0.9077), prey taxa
Finless porpoises in western Kyushu fed on various species
of fishes, cephalopods, and crustaceans. These findings are
consistent with those observed in Hong Kong waters (Barros
et al. 2002) and Korean waters (Park et al. 2005). However,
prey composition of the diets differed among the areas. In the
Ariake Sound–Tachibana Bay area, cephalopods were the most
numerous component. In contrast, fish predominated in Omura
Bay and Hong Kong waters (Barros et al. 2002), and
crustaceans predominated in Korea (Park et al. 2005).
Geographical variation in diet may reflect differences in
availability and selectivity of prey. Here we focus on differences in prey availability between the Ariake Sound–Tachibana
Bay area and Omura Bay.
Ariake Sound surpasses Omura Bay with respect to species
richness of fishes and cephalopods. Approximately 250 species
of fish and 13 species of cephalopods may occur in Ariake
Sound (Kikuchi 2001), whereas 107 species of fish and 10
species of cephalopods are confirmed in Omura Bay (Takita
1985). A geographical difference in the number of prey items
detected (40 prey items in the Ariake Sound–Tachibana Bay
area and 12 in Omura Bay) may reflect not only a difference in
the sample size of the porpoises but also the difference in
species richness. For example, a green eelgoby, which is
endemic to the sound in Japan (Takita 2001), was found in
a stomach of a porpoise. Commercial catch statistics of fish,
cephalopods, and shrimp indicated that fish were dominant in
the catch in Ariake Sound, Tachibana Bay, and Omura Bay
(Table 4), but species compositions differed: the fish species
with the dominant catch was konoshiro gizzard shad in Ariake
Sound and Japanese pilchard together with Japanese anchovy
in Tachibana Bay. Cephalopods were likely to be more
abundant than shrimp in Ariake Sound, whereas shrimp was
likely to be abundant in Omura Bay, with the highest
percentage of 13.6% of total catch. These differences may
explain the geographical difference in diet between populations. In Omura Bay, shoals of Gobiidae have been observed in
the marginal areas of oxygen-deficient waters that occur
frequently (Takita 1985). The high %N of Gobiidae in the
bay may be attributed to a high level of availability of this fish.
Some prey selectivity exists. Sea breams were commercially
important, with a high level of catch in Ariake Sound and Omura
Bay (Table 4), but they were not confirmed in stomachs of finless
porpoises. Other evidence is given by our questionnaire survey
of the fishermen who provided the specimens of the porpoises.
Of the species captured by fishing nets in which the porpoises
were entangled, gurnard, Japanese whiting, sea bream, bastard
halibut, flounder, tonguefish, filefish, marbled rockfish, puffer
fish, and crabs were not confirmed in the stomachs.
Shirakihara et al. (1993) indicated the parturition season of
finless porpoises differed between western Kyushu areas and
other Japanese waters of the Inland Sea and the Ise Bay–
Mikawa Bay area. Because they did not specify the season by
1254
Vol. 89, No. 5
JOURNAL OF MAMMALOGY
TABLE 4.—Commercial catch in Ariake Sound, Tachibana Bay, and Omura Bay cited from annual statistics of fisheries in Nagasaki,
Kumamoto, Saga, and Fukuoka prefectures surrounding Ariake Sound, Tachibana Bay, or Omura Bay in 1987–1992.
Fisheries resources
Fish
Total
a
b
c
Tons
Tons
%
81.3
141
22
11,687
4,100
4,851
9,042
5,325
374
34,825
11,390
Octopoids
Cuttlefish
Others
Shrimp
Tachibana Bay (700 km2)
70,367
Japanese pilchard
Japanese anchovy
Konoshiro gizzard shadb
Mullet
Croaker
Flounder/halibut
Sea bream
Japanese horse mackerel
Others
Cephalopodsc
Ariake Sound (1,700 km2)
13.2
5,960
3,873
1,557
%
150,018
0.2
0.0
13.5
4.7
5.6
10.4
6.2
0.4
40.2
Tons
97.5
82,270
33,836
—
180
—
433
898
4,760
27,641
1,676
6.9
4.5
1.8
Omura Bay (320 km2)
%
9,099
1.1
562
496
618
76.4
877a
773a
—
245
—
338
1,201
245
5,420
53.5
22.0
—
0.1
—
0.3
0.6
3.1
18.0
1,191
0.4
0.3
0.4
7.4
6.5
—
2.1
—
2.8
10.1
2.1
45.5
10.0
455
270
466
3.8
2.3
3.9
4,791
5.5
2,113
1.4
1,625
13.6
86,548
100.0
153,807
100.0
11,915
100.0
Catch of Japanese anchovy in Sumo Nada was excluded.
Catch was underbiased because ‘‘konoshiro gizzard shad’’ was regarded as ‘‘other fishes’’ in the statistics of some prefectures.
Catch of Japanese common squid was excluded because its main fishing ground was in East China Sea.
population in western Kyushu, we reanalyzed the data they
used. The parturition season for the Ariake Sound–Tachibana
Bay population was from August to March with a peak in
November–December and an indistinct peak in March. The
parturition season for the Omura Bay population was not clear
because of a small sample size, but examination of aerial
sighting data of mother and calf pairs indicates that the season
may be from spring to summer (K. Shirakihara, in litt.). In
Inland Sea and the Ise Bay–Mikawa Bay area, the parturition
season is from March to August with a distinct peak in April
(Furuta et al. 1989; Kasuya and Kureha 1979).
Calves born in the fall–winter season will be taking solid
food in spring to summer in the Ariake Sound–Tachibana Bay
area, because mean length at weaning of 101 cm (95% CI ¼
95–105 cm) corresponds to approximately 6 months (4–8
months) of age according to the growth equation given by
Shirakihara et al. (1993). This is almost consistent with the
observation of finless porpoises in captivity: 2 calves that were
considered to be weaned smoothly began to feed on solid food
at 120 and 132 days of age (Furuta 2008; M. Furuta, Toba
Aquarium, pers. comm.). Finless porpoises began to take solid
food at 6–12 months of age in Hong Kong waters (Barros et al.
2002). Kasuya and Kureha (1979) estimated that finless
porpoises may be weaned at ages ranging from 6 to 15 months
and commonly have a 2-year breeding cycle. Reproduction
may be related to seasonal availability of important food items
for cetacean mothers or weaning calves if there are substantial
seasonal changes in prey availability (Whitehead and Mann
2000). Harbor porpoise (Phocoena phocoena) calves are
considered to be born when maternal prey is abundant and of
high quality (Börjesson and Read 2003). The prey of shortfinned pilot whales (Globicephala macrorhynchus) appears to
be abundant when most calves are expected to wean (Kasuya
and Tai 1993). Birth in the Ariake Sound–Tachibana Bay area
takes place mainly in fall and winter, when the water
temperature is decreasing. Although monthly catch statistics
are undisclosed, reported prey occurrence or distribution
showed that there may be seasonal changes of prey availability
in the Ariake Sound–Tachibana Bay area. Konoshiro gizzard
shad appears to be an important prey for lactating females as
well as other individuals. This species, which inhabits the
sound all year and lives for at least 3 years (Takita 1978),
showed high levels of %N and %F in the fall–winter season.
Few finless porpoises were sighted in shallow waters , 5 m
deep in the sound in daylight hours (Yoshida et al. 1997).
Similarly low numbers of sightings in waters , 10 m deep also
were reported for other Japanese waters (Amano et al. 2003;
Shirakihara et al. 2007). A low contribution of konoshiro
gizzard shad as prey in spring to summer may be related to its
migration to shallow waters in spring to fall (Takita 1978;
Takita et al. 2003). The main weaning season of spring to
summer almost coincides with the spawning seasons of prey of
weaning calves. Gobiidae and most of the cephalopods spawn
from spring to summer and from early spring to early fall,
respectively (Takita 1980; Y. Natsukari, Nagasaki University,
pers. comm.). Although it is unknown whether availability of
a prey species becomes high during its spawning season, Hirai
and Nishinokubi (2004) reported that small cuttlefish (mantle
length , 9.0 cm) accounted for 52.6% of the total catch by
small-scale trawlers in February in Ariake Sound. In addition,
cuttlefish appear to spawn 3 or 4 months earlier than that in the
Inland Sea, Mikawa Bay or Tokyo Bay, and in January to
June in Ariake Sound (Watanuki and Kawamura 1999). The
fall–winter season of birth of finless porpoises may be related
October 2008
SHIRAKIHARA ET AL.—FOOD HABITS OF FINLESS PORPOISES
to the seasonal availability of shad or cephalopods in the
Ariake Sound–Tachibana Bay area.
ACKNOWLEDGMENTS
We thank T. Takita, Y. Natsukari, and K. Mizuta for useful
suggestions about fish and cephalopods in our study area and for
identifying prey items; O. Tabeta and A. Tamaki for helpful
suggestions in identifying prey items; F. Ohe for identifying the
otoliths that were not included in our collection; members of the
Laboratory of Fisheries Science, Nagasaki University, for their
assistance in collecting biological samples; M. Yoshioka and H.
Ohizumi for their valuable comments and helpful discussions; M.
Furuta and T. Kasuya for their encouragement during the study; and
W. F. Perrin, E. Heske, and 2 anonymous reviewers for invaluable
comments and useful suggestions. This study was funded by a grant
from Fujiwara Natural History Foundation.
LITERATURE CITED
AMANO, M. 2002. Finless porpoise Neophocaena phocaenoides. Pp.
432–435 in Encyclopedia of marine mammals (W. F. Perrin, B.
Würsig, and J. G. M. Thewissen, eds.). Academic Press, San Diego,
California.
AMANO, M., F. NAKAHARA, A. HAYANO, AND K. SHIRAKIHARA. 2003.
Abundance estimate of finless porpoises off the Pacific coast of
eastern Japan based on aerial surveys. Mammal Study 28:103–110.
ARCHER, F. I., AND K. M. ROBERTSON. 2004. Age and length at
weaning and development of diet of pantropical spotted dolphins,
Stenella attenuata, from the eastern tropical Pacific. Marine
Mammal Science 20:232–245.
BARROS, N. B., T. A. JEFFERSON, AND E. C. M. PARSONS. 2002. Food
habits of finless porpoises (Neophocaena phocaenoides) in Hong
Kong waters. Raffles Bulletin of Zoology Supplement 10:115–123.
BÖRJESSON, P., AND A. J. READ. 2003. Variation in timing of
conception between populations of the harbor porpoise. Journal of
Mammalogy 84:948–955.
CHEN, P., P. LIU, R. LIU, K. LIN, AND G. PILLERI. 1979. Distribution,
ecology, behaviour and conservation of the dolphins of the middle
reaches of Changjiang (Yangtze) River (Wuhan-Yueyang). Investigations on Cetacea 10:87–103.
CLARKE, M. R. 1986. A handbook for the identification of cephalopod
beaks. Clarendon Press, Oxford, United Kingdom.
FURUTA, M. 2008. Breeding and reproduction in captive finless
porpoise. Aquabiology 30:38–43.
FURUTA, M., T. KATAOKA, M. SEKIDO, K. YAMAMOTO, O. TSUKADA,
AND T. YAMASHITA. 1989. Growth of the finless porpoise Neophocaena phocaenoides (G. Cuvier, 1829) from the Ise Bay,
Central Japan. Annual Report of Toba Aquarium 1:89–102.
GANNON, W. L., R. S. SIKES, AND THE ANIMAL CARE AND USE
COMMITTEE OF THE AMERICAN SOCIETY OF MAMMALOGISTS. 2007.
Guidelines of the American Society of Mammalogists for the use of
wild mammals in research. Journal of Mammalogy 88:809–823.
HIRAI, Y., AND H. NISHINOKUBI. 2004. By-catch and discards of
marketable species for small-scale trawler in Ariake Sea. Nippon
Suisan Gakkaishi 70:738–744.
INOUE, N. 1980. Physical environment in Ariake Sound. Kaiyo
Monthly 12:116–126.
KASUYA, T., AND K. KUREHA. 1979. The population of finless porpoise
in the Inland Sea of Japan. Scientific Reports of the Whales
Research Institute 31:1–44.
KASUYA, T., AND S. TAI. 1993. Life history of short-finned pilot whale
stocks off Japan and a description of the fishery. Pp. 439–473 in
1255
Biology of Northern Hemisphere pilot whales (G. P. Donovan,
C. H. Lockyer, and A. R. Martin, eds.). International Whaling
Commission, Cambridge, United Kingdom.
KATAOKA, T., H. KITAMURA, M. SEKIDO, AND K. YAMAMOTO. 1976. On
the feeding habit of finless porpoise (Neophocaena phocaenoides).
Journal of Japanese Association of Zoological Gardens and
Aquariums 18:49–56.
KIKUCHI, T. 2001. Conservation of tidal flats and estuaries. Pp. 306–
317 in Life in Ariake Sea: biodiversity in tidal flats and estuaries
(M. Sato, ed.). Kaiyu-sha, Tokyo, Japan.
MASAKI, Y. 1980. On the finless porpoises taken in Suruga Bay.
Geiken-Tsushin 335:35–40.
MASUDA, H., K. AMAOKA, C. ARAGA, T. UENO, AND T. YOSHINO. 1984.
Reference of Japanese fishes. Tokai University Press, Kanagawa,
Japan.
MATSUBARA, K. 1979. Morphology and key to species for fishes.
Ishizaki-shoten, Tokyo, Japan.
OHE, F. 1985. Marine fish-otoliths of Japan. Senior High School
Attached to the Aichi University of Education, Aichi, Japan.
PARK, K. J., C. I. ZHANG, H. SOHN, AND Z. G. KIM. 2005. Feeding
habits of finless porpoise (Neophocaena phocaenoides) in the west
coast of Korea. Paper SC/57/SM17. IWC, Impington, Cambridge,
United Kingdom.
PILLERI, G., AND M. GIHR. 1972. Contribution to the knowledge of
cetaceans of Pakistan with particular reference to the genera
Neomeris, Sousa, Delphinus and Tursiops and description of a new
Chinese porpoise (Neomeris asiaeorientalis). Investigations on
Cetacea 4:107–162.
SHIRAKIHARA, K., M. SHIRAKIHARA, AND Y. YAMAMOTO. 2007.
Distribution and abundance of finless porpoise in the Inland Sea
of Japan. Marine Biology 150:1025–1032.
SHIRAKIHARA, K., H. YOSHIDA, M. SHIRAKIHARA, AND A. TAKEMURA.
1992a. A questionnaire survey on the distribution of the finless
porpoise, Neophocaena phocaenoides, in Japanese waters. Marine
Mammal Science 8:160–164.
SHIRAKIHARA, M., K. SHIRAKIHARA, AND A. TAKEMURA. 1992b. Records
of the finless porpoise (Neophocaena phocaenoides) in the waters
adjacent to Kanmon Pass, Japan. Marine Mammal Science 8:82–85.
SHIRAKIHARA, M., K. SHIRAKIHARA, AND A. TAKEMURA. 1994.
Distribution and seasonal density of the finless porpoise Neophocaena phocaenoides in the coastal waters of western Kyushu,
Japan. Fisheries Science 60:41–46.
SHIRAKIHARA, M., A. TAKEMURA, AND K. SHIRAKIHARA. 1993. Age,
growth, and reproduction of the finless porpoise, Neophocaena
phocaenoides, in the coastal waters of western Kyushu, Japan.
Marine Mammal Science 9:392–406.
TAKAHASHI, Z. 1962. Study for the identification of species based on
the vertebral column of Teleostei in the Inland Sea and its adjacent
waters. Bulletin of Naikai Regional Fisheries Research Laboratory
16:1–71.
TAKITA, T. 1978. Reproductive ecology of a shad, Konosirus
punctatus, in Ariake Sound—I. Distribution, body condition, and
maturation. Bulletin of the Faculty of Fisheries, Nagasaki
University 45:5–10.
TAKITA, T. 1980. Fishes in Ariake Sound. Marine Science Monthly
12:105–115.
TAKITA, T. 1985. Fauna in Omura Bay. Pp. 894–900 in Coastal
oceanography of Japanese Islands Tokai University Press, Tokyo,
Japan.
TAKITA, T. 2001. Fishes. Pp. 213–252 in Life in Ariake Sea:
biodiversity in tidal flats and estuaries (M. Sato, ed.). Kaiyu-sha,
Tokyo, Japan.
1256
JOURNAL OF MAMMALOGY
TAKITA, T., D. KOMURA, I. KAWAHARA, Y. MORI, N. NAKASHIMA, AND
S. ITO. 2003. Distribution of fishes in the innermost area of Ariake
Sound. Bulletin of Saga Prefectural Ariake Fisheries Research and
Development Center 21:81–98.
WATANUKI, N., AND G. KAWAMURA. 1999. A review of cuttlefish
basket trap fishery. South Pacific Study 19:31–48.
WHITEHEAD, H., AND J. MANN. 2000. Female reproductive strategies of
cetaceans life histories and calf care. Pp. 219–246 in Cetacean
societies field studies of dolphins and whales (J. Mann, R. C.
Connor, P. L. Tyack, and H. Whitehead, eds.). University of
Chicago Press, Chicago, Illinois.
XIAO, J. Q., K. X. WANG, AND D. WANG. 2005. Diurnal changes of
behavior and respiration of Yangtze finless porpoises (Neophocaena
phocaenoides asiaeorientalis) in captivity. Zoo Biology 24:
531–541.
Vol. 89, No. 5
YOSHIDA, H., K. SHIRAKIHARA, H. KISHINO, AND M. SHIRAKIHARA. 1997.
A population size estimate of the finless porpoise, Neophocaena
phocaenoides, from aerial sighting surveys in Ariake Sound and
Tachibana Bay, Japan. Researches on Population Ecology 39:239–
247.
YOSHIDA, H., M. YOSHIOKA, M. SHIRAKIHARA, AND S. CHOW. 2001.
Population structure of finless porpoises (Neophocaena phocaenoides) in coastal waters of Japan based on mitochondrial DNA
sequences. Journal of Mammalogy 82:123–130.
ZHOU, K., A. GAO, AND J. SUN. 1993. Notes on the biology of the
finless porpoise in Chinese waters. IBI Reports 4:69–74.
Submitted 29 August 2007. Accepted 19 February 2008.
Associate Editor was William F. Perrin.