Gannon et al.: Food habits of Phocoena phocoena 
431 
trieved from skulls) to 5 (severely degraded, free 
otoliths) following the methods of Recchia and Read 
(1989). Otoliths categorized as 3 or higher were not 
used in size estimations, unless no undamaged 
otoliths were present. When only damaged otoliths 
from a particular prey species were present in a por- 
poise stomach, the available skeletal structures were 
measured; consequently the reconstituted prey mass 
for that stomach may have been underestimated (see 
Jobling and Breiby, 1986; Sekiguchi and Best, 1997) 
These three measures of prey importance were 
applied to data from the 82 noncalf porpoises as a 
group and to each sex and maturity class. Food habit 
studies in which different methods are used can yield 
widely disparate results, making it difficult to draw 
comparisons between studies (Gannon et al., 1997a, 
1997b). Because one of the primary objectives of this 
research was to obtain information on seasonal 
changes in the diet, it was important for these data 
to be treated in a manner similar to those of Recchia 
and Read (1989) and Smith and Read (1992). 
Results 
Overall sample 
Table 2 lists the numbers and mean sizes of 15 prey 
taxa recovered from the 95 porpoise stomachs. At- 
lantic herring ( 78%), silver hake ( Merluccius bilinearis, 
68%), pearlsides ( Maurolicus weitzmani, 38%), and red 
and white hake ( Urophycis spp., 29%) occurred most 
frequently in the stomachs of the 74 noncalf porpoises 
(Table 3). Atlantic herring represented only 7% of 
the food by proportion of numerical abundance but 
accounted for 44% of ingested mass. Pearlsides ac- 
counted for 67% of food by proportion of numerical 
abundance but only 3% by ingested mass, owing to 
their small size. The unknown fish present in por- 
poise stomachs may have been alewives ( Alosa 
pseudoharengus ) but this could not be determined 
with certainty. Both red and white hake ( Urophycis 
chuss and U. tenuis ) were present; however it is dif- 
ficult to differentiate between small, eroded otoliths 
from red and white hake, therefore all Urophycis 
otoliths were grouped together. Atlantic hagfish 
(Myxine glutinosa) and euphausiids ( Meganycti - 
phanes norvegica ) were included in analyses of fre- 
quency of occurrence only because the numerical 
abundance and mass of these two species were diffi- 
cult to estimate. To allow comparisons to be drawn 
with the summer diet, data from Recchia and Read 
(1989) are also given in Table 3. 
Figure 2 shows length-frequency distributions for the 
three most abundant prey: pearlsides, silver hake, and 
Atlantic herring. On average, Atlantic herring was the 
largest prey consumed by length (254 mm ±36 SD ) with 
a range from 159 to 339 mm. The average fork length 
Table 2 
Number and mean sizes of food items present in the stomachs of harbor porpoises sampled in the Gulf of Maine during autumn. 
ML = mantle length, FL = fork length, and SL = standard length. Present = present in porpoise stomach contents but numerical 
abundance not determined. 
Food item 
n 
Length 
measurement 
Mean 
length ± SD 
(mm) 
Mean 
mass ± SD 
<g) 
Bathypolypus arcticus 
i 
ML 
52 
48 
Clupea harengus 
507 
FL 
254 + 36 
133 ± 56 
Gadus morhua 
5 
FL 
241 ± 133 
137 ± 201 
Illex illecebrosus 
18 
ML 
— 
55 ± 22 
Loligo pealei 
8 
ML 
129 ± 30 
68 + 29 
Maurolicus weitzmani 
5898 
FL 
50 ± 4 
0.9 ± 0.2 
Meganyctiphanes norvegica 
present 
— 
— 
— 
Merluccius bilinearis 
1605 
FL 
164 ± 96 
65 ± 88 
Myxine glutinosa 
present 
— 
— 
— 
Peprilus triacanthus 
38 
SL 
97 ± 12 
24 ± 7 
Pollachius virens 
76 
FL 
195 ± 101 
136 ± 130 
Scomber scombrus 
15 
FL 
224 ± 53 
127 ± 91 
Sebastes spp. 
47 
FL 
37 + 3 
0.7 ± 0.2 
Urophycis spp. 
474 
FL 
159 ± 146 
111 ± 172 
Unknown fish 
4 
— 
— 
— 
Milk 
present 
— 
— 
— 
