Gannon et al.: Food habits of Phocoena phocoena 
435 
cation that some porpoise calves begin weaning 
sooner than others. The species composition found 
in the stomachs of calves in autumn begins to re- 
semble that of older animals. However, the propor- 
tions of prey types and sizes of prey differ from those 
of adults. In autumn, calves eat a greater proportion 
of pearlsides and euphausiids than do older animals, 
and the sizes of Atlantic herring and silver hake are 
smaller than those eaten by older porpoises. Pearl- 
sides, euphausiids, juvenile silver hake, juvenile her- 
ring, and juvenile red and white hake appear to be 
important in the “transitional diet” of calves, as they 
learn to forage independently. Calves eat a larger 
quantity and greater diversity of solid food in au- 
tumn than in the summer (Smith and Read, 1992). 
Our observations support and extend the findings of 
Smith and Read ( 1992), who suggested that porpoise 
calves eat euphausiids while their mothers are feed- 
ing on other euphausiid predators. 
Although harbor porpoises prey on some of the 
groundfish species targeted by the sink gillnet fish- 
ery in the Gulf of Maine, these species contribute 
just a small fraction of the overall diet. Furthermore, 
the size range of groundfish consumed by porpoises 
is much smaller than that targeted by the gillnet fish- 
ery because porpoises feed on only the juvenile age 
classes of those commercial species. The prey that 
represent the bulk of the porpoise diet (i.e. Atlantic 
herring, silver hake, and pearlsides) are important 
forage items for groundfish targeted by the sink 
gillnet fishery (Langton, 1982). These dietary simi- 
larities may lead to overlap between the distribu- 
tions of groundfish and porpoises, leading both to be 
caught in the same nets. Silver hake found in por- 
poise stomachs were highly digested (only 0.1% of 
silver hake were intact), indicating that they had 
been consumed some time prior to entanglement. In 
contrast, herring were often found in a relatively 
undigested state (15.8% were intact), indicating that 
many porpoises had been feeding on herring at, or 
just before, the time of entanglement. 
Several potential biases should be kept in mind 
when interpreting these results. First, all the por- 
poises we examined had been killed in gill nets an- 
chored to the ocean floor. This capture method may 
have led to a bias towards demersal prey and against 
pelagic prey. Without comparable samples collected 
near the surface, it is not possible to fully address 
this potential bias. The samples of Recchia and Read 
(1989) and Smith and Read (1992) may be similarly 
biased because both studies also obtained samples 
from porpoises killed in sink gill nets. Second, dif- 
ferential digestion and retention of hard parts are 
unavoidable in studies of marine mammal stomach 
contents. Consequently, the importance of species 
that are resistant to digestion, or that accumulate 
in porpoise stomachs, will be overestimated. With- 
out empirical data on digestion times for each prey 
species, it is not possible to evaluate this potential 
bias fully. 
A third potential source of bias arises from the dif- 
ficulty in discriminating between primary prey (con- 
sumed by porpoises) and secondary prey (consumed 
by porpoise prey). For example, it is possible that 
small organisms, such as pearlsides, euphausiids, 
and juvenile silver hake, were secondarily introduced 
into the porpoise stomach contents. Careful exami- 
nation of species co-occurrences in porpoise stomachs 
can provide insights into whether these small organ- 
isms were actually eaten by the porpoises. Because 
many porpoise prey are euphausiid predators 
(Bigelow and Schroeder, 1953; Langton, 1982; Scott 
and Scott, 1988), it is difficult to evaluate the likeli- 
hood of secondary consumption of euphausiids. How- 
ever, two calves had euphausiid remains but no other 
solid food in their stomachs, indicating that they had 
consumed the euphausiids directly. One calf had 
pearlsides remains and a herring in its stomach; 
herring are not considered predators of pearlsides 
(Bigelow and Schroeder, 1953; Scott and Scott, 1988). 
Five calves had remains of pearlsides together with 
juvenile red, white, and silver hake less than 57 mm 
in length, too small to be predators of pearlsides. We 
interpret the co-occurrence of pearlsides and juve- 
nile gadiforms in stomachs of calves as an indication 
of their preference for small prey, rather than as the 
presence of predators and secondary prey in their 
stomachs. Among older porpoises, one individual had 
pearlsides with no other food remains; four had 
pearlsides and herring; one had 13 pearlsides (total- 
ing 16 grams), a 14-gram butterfish, and a herring; 
and one had 1100 pearlsides (1052 g) and one but- 
terfish (6 g). Therefore, it is apparent that porpoises 
do indeed prey directly on euphausiids, pearlsides, 
and juvenile gadiforms. 
In conclusion, the seasonal movements of harbor 
porpoises are accompanied by changes in diet. Sea- 
sonal movements of porpoises may, in fact, be driven 
by their need to maintain proximity to sufficient con- 
centrations of prey. Assuming that there have not 
been any major shifts in prey availability between 
the previous study in the Bay of Fundy (Recchia and 
Read, 1989) and the present study, the diet of har- 
bor porpoises in the Gulf of Maine during autumn 
appears to be more diverse than that of harbor por- 
poises in the Bay of Fundy during summer. The win- 
ter ecology of this population probably differs also 
because many porpoises are believed to leave the Gulf 
of Maine and Bay of Fundy region during this sea- 
son. Further information on the diet of this popula- 
