Carrasson and Matallanas: Feeding ecology of Bathypterois mediterraneus 
267 
Sampling data for Bathypterois mediterraneus 
of specimens sampled. 
Table 1 
in the present study. N lat = 
North latitude; E long 
= East longitude, n 
= number 
Station 
Date 
Depth (m) 
(initial final) 
Final situation 
n 
N lat 
E long 
BII-4 
Jul 29 1987 
1432 1419 
40 44.7 
1 52.6 
2 
BII-5 
Jul 30 1987 
1753 1715 
40 25.4 
1 56.9 
60 
BII-6 
Jul 31 1987 
1287 1329 
40 54.7 
2 11.5 
55 
BII-8 
Aug 1 1987 
1295 1357 
41 02.6 
2 27.8 
59 
BIII-3 
Jun 25 1988 
1774 1783 
40 18.5 
1 57.2 
33 
BIII-4 
Jun 26 1988 
2163 2039 
40 37.7 
3 06.2 
16 
BIII-5 
Jun 26 1988 
2256 2239 
40 32.3 
3 44.7 
80 
The object of this paper is to provide new and detailed 
information on the diet of Bathypterois mediterraneus 
from depths of 1000-2250 m in the western Mediterra- 
nean. Its feeding habits are analyzed in different bathy- 
metric strata and within groups of immature and adult 
specimens. The influence of these factors on diet, and pat- 
terns of dietary overlap, are also discussed. 
Materials and methods 
Samples were collected from the continental slope 
(1000-2250 m) of the Catalan Sea (western Mediterra- 
nean), during two cruises (BATHOS II— III ), on board the 
RV Garcia del Cid (Table 1), with a semi-balloon otter- 
trawl (OTSB14) towed from a single warp (cf. Merret and 
Marshall, 1981). 
All specimens were fixed in 10% formalin immediately 
after capture. Once in the laboratory, they were measured 
(standard length: SL, to the nearest millimeter) and dis- 
sected to analyze the gut contents. A total of 305 speci- 
mens were dissected to examine feeding activity. 
The linear and poorly differentiated stomach of Bathyp- 
terois mediterraneus is almost always found empty; prey 
are usually found in the intestine. Therefore, intestinal 
contents were analyzed, which makes identifying prey 
more difficult, because of advanced digestion. 
Food items were identified to the lowest taxonomic lev- 
el possible. Numbers and weights were registered to the 
nearest 0.1 mg, after items were dried with blotting paper 
to remove surface moisture. 
The quantitative importance of each prey group in the 
diet was determined by the index of relative importance 
(IRI) (Pinkas et ah, 1971), defined as 
IRI = %F(%N + %V), 
where %F = frequency of occurrence of the food item; 
%N = numerical percentage of a food item in the 
stomachs; and 
%V = percentage by volume of the food item in the 
stomachs (Hureau, 1970). 
In our study, weight was used (%W) instead of volume 
(%M). This modified index has been expressed as %IRI - 
(IRI/URI) x 100 (Rosecchi and Nouaze, 1987). 
To analyze the diet of Bathypterois mediterraneus, indi- 
viduals from all trawls were grouped according to capture 
depth (three bathymetric strata: 1000-1425 m; 1425-1800 
m; and 1800-2250 m) and size of individuals (two catego- 
ries: immature or size 1, standard length <113 mm; and 
mature or size 2, standard length >113 mm). The %IRI of 
the main prey items was determined for each of the six 
combinations of depth and size by pooling diet data from 
the individuals included in each combination. The affinity 
of these six combinations was computed by using a hier- 
archical analysis (weighted-pair groups methods analysis, 
WPGMA). 
Trophic diversity (H') was calculated, in terms of mean 
%W of prey items, by using the Shannon index. Degree 
of overlap in the diet of Bathypterois mediterraneus, by 
different sizes and bathymetric strata, was determined, 
based on mean %W results, by using the quantitative 
Schoener index (Schoener, 1974). 
Results 
Of the 305 specimens of Bathypterois mediterraneus ana- 
lyzed, 23 had an empty gut. Forty-nine categories of 
prey items were identified from the 282 guts containing 
food (Table 2). Calanoid copepods were the most numeri- 
cally abundant prey (%N=81. 1 1%), and mysids were the 
most abundant by mass (%W=37.59%). Calanoid cope- 
pods (%IRI=91. 63), according to Wishner ( 1980) and Smith 
( 1982 ), are planktonic elements of the benthopelagic fauna. 
Benthic fauna are accidental prey. 
From the cluster analysis of the size and depth combina- 
tions, we identified four groups (Fig. 1): juveniles collected 
at depths of 1800-2250 m (group A), adults collected at 
