232 
Fishery Bulletin 99(2) 
Kogia spp. 
The distributions of the Kogia 
spp. with respect to depth, 
depth gradient, and epipelagic 
zooplankton biomass were sig- 
nificantly different from a uni- 
form distribution (Table 5); 
however, the distribution with 
respect to depth gradient may 
have been an artifact of the 
depth distribution (P>0.05; 
Table 5). The distribution with 
respect to surface temperature 
variability was significantly dif- 
ferent from the effort (P<0.05; 
Table 5), and the median value 
of surface temperature vari- 
ability for Kogia spp. was the 
highest of all the species exam- 
ined (Fig. 5D). Kogia spp. had 
a modal distribution about the 
upper continental slope (Fig. 
9A) and group sighting rates 
increased with increasing zoo- 
plankton biomass (Fig. 9B). 
A bivariate x 2 analysis (after 
Baumgartner, 1997) indicated 
that the distribution of Kogia 
spp. was significantly different 
from a uniform distribution 
with respect to both depth and 
zooplankton biomass (^ 2 — 29.2, 
df=4, P<0.0001) and that in 
waters of high zooplankton bio- 
mass over the upper continen- 
tal slope, group sighting rates 
were 2.5 times the average. 
Pantropical spotted dolphin 
The distribution of the pan- 
tropical spotted dolphin was 
significantly different from a 
uniform distribution of depth, 
depth gradient, and surface 
chlorophyll (Table 5); however, 
the distributions with respect 
to depth gradient and surface 
chlorophyll may have been an artifact of the depth distri- 
bution (P>0.05;Table 5). This species was encountered only 
once near the shelf break (Fig. 10) and if the x 2 analysis 
was limited to sightings and effort deeper than 500 m, then 
the depth distribution would not be significantly different 
from a uniform distribution (/ 2 =14.5, df=ll, P>0.05). 
Sperm whale 
The distribution of the sperm whale with respect to depth 
was significantly different from a uniform distribution 
(Table 5; Fig. 11A). Only one group of sperm whales was 
encountered near the shelf break and if the x 2 analysis 
was limited to sightings and effort deeper than 500 m, 
then the depth distribution would not be significantly dif- 
ferent from a uniform distribution (j 2 =2.2, df=7, P>0.05). 
There is evidence to suggest that the distribution with 
respect to the depth of the 15°C isotherm was significantly 
different from a uniform distribution when the depth dis- 
tribution is taken into consideration (Monte Carlo test, 
P<0.05; Table 5). In waters where the 15°C isotherm was 
deeper than 200 m, the group sighting rate of sperm 
