668 
Fishery Bulletin 96(4), 1998 
(Bonnaterre, 1788), and the requiem sharks, Carchar- 
hinus spp. (Garces and Rey, 1984). Bony fish are 
caught occasionally, including the northern bluefin 
tuna, Thunnus thynnus (Linnaeus, 1758), bigeye 
tuna, Thunnus obesus (Lowe, 1839), albacore tuna, 
Thunnus alalunga (Bonnaterre, 1788), oilfish, 
Ruvettus pretiosus Cocco, 1829, escolar, Lepido- 
cybium flavobrunneum (Smith, 1849), and longbill 
spearfish, Tetrapturus pfluegeri Robins and de Sylva, 
1963, (Mejuto and Garces 1 ; Mejuto 2 ). 
Materials and methods 
Between July 1991 and July 1992, 106 longline and 
69 gillnet landings (85.4% and 87.3% respectively of 
total longline and gillnet landings) were sampled at 
the Algeciras fish market (Cadiz, southern Spain), 
the largest market in southwestern Spain receiving 
80% of the longline catch from this area (Anonymous, 
1986). Data on the number of fish landed by species, 
days of each trip, active fishing days, number of sets 
and hooks for longliners, and net length and fishing 
time for gillnetters were collected during interviews 
with skippers. 
The longline catch rate was calculated as hook rate 
(HR=number of fish/1,000 hooks), and gillnet catch 
rate was calculated as the average net length by trip 
multiplied by the number of sets. No total catch and 
effort data were available; catch rates are approxi- 
mate indicators and do not reflect actual abun- 
dance. Fishing time was not included in effort 
calculations because it was constant for every 
boat and gear. The gears were set at dusk and 
retrieved before sunrise. Longlines ranged from 
18 to 29 km in length and on average had 1500 
hooks (range: 475 to 2500), set at a depth of 11 
to 55 m. Gill nets were 2.5 km long, 14 m high, 
and had a mesh size of 40 cm. 
All fish landed were identified to species level 
and counted. Most sharks were sexed and mea- 
sured to the nearest centimeter for total length 
(TL) or fork length (FL), or both. Total length 
was measured from the rear tip of the upper cau- 
dal lobe to the snout tip, along the horizontal line 
of the body axis. Length data were grouped in 5- 
cm intervals of FL (shortfin mako and blue sharks) 
and TL (thresher and hammerhead sharks) for 
length-frequency analysis. Biometrical relation- 
ships between sexes were performed by simple 
linear regression analysis when the sample size 
was sufficiently large (shortfin mako, blue, and 
bigeye thresher sharks). Standard length (SL) has 
been used as the reference length for the relation 
with fork length (FL), total length (TL), and 
upper caudal lobe (UCL) (dorsal-caudal margin). The 
relation between FL and TL was also calculated. The 
area sampled (Fig. 1) was divided into 5° latitude 
sectors from 20°N to 40°N. Sectors 4 and 5 have the 
same latitude (35°N-40°N) but different longitudes, 
3°W-8°W for sector 5, 8°W-13°W for sector 4. This 
division separates the gillnet fishing ground (sector 
5) from the longline Atlantic area (sector 4). 
A chi-square (% 2 ) test was performed to test the fit 
of the sample to a normal distribution and to observe 
the variations in species by sector and month. Strong 
bias was expected because of the effect of discarded 
fish at sea. Single factor AN OVA was performed to 
compare the means of the samples by area. 
Results 
A total of 40,198 sharks and 11,007 bony fish were 
sampled from 175 landings from July 1991 to July 
1992 (Table 1). The proportion of sharks landed was 
higher with longlines than with gill nets, but this 
might be related to the seasonal abundance of sharks 
and swordfish in the Strait of Gibraltar during the 
period when gill nets were used (Fig. 2). 
The most common species landed by longlines in 
the areas studied were blue shark, shortfin mako 
shark, and swordfish (Table 2). Other chondrich- 
thyans were less common, some rarely occurring in 
more than 10% of the landings. 
