Eggs and Larvae of the Nehu — TESTER 
335 
Fig. 3. A portion of the southern sector of Kaneohe 
Bay in the vicinity of Stations 4 and 5, showing the 
location of A and B hauls in Survey \b, and indicating 
the presumed distribution of the eggs (broken iso- 
lines). 
cause of the small numbers of larvae taken in 
the samples, a similar detailed analysis of the 
data was not undertaken. 
Obviously, the peculiar non-random distri- 
bution of the eggs should be taken into con- 
sideration in designing a program of quan- 
titative sampling for future work. 
VERTICAL DISTRIBUTION OF EGGS AND LARVAE 
The vertical distribution of eggs and larvae 
was studied only at Station 4 during Surveys 
la and lb, the details of which have been dis- 
cussed in the previous section. Unfortunately, 
it was not possible to sample efficiently over a 
range of depth. In analyzing the results 
(Tables 9 and 11), the surface samples, from 
1 to 2 feet, are compared with those at *'a 
depth,” the depth ranging from 6 to 17 feet 
in Survey la (Table 8) and from 16 to 20 
feet in Survey lb (Table 10). The (geometric) 
mean catches of eggs in the adjusted data may 
be compared in the following tabulation: 
HAULS SURVEY la SURVEY Ih 
Surface 134 54 
Deep 107 41 
In both surveys there were more eggs at 
the surface than at a depth. There were no 
significant interactions involving depth in 
either Survey la (RXD and HXD) or Sur- 
vey lb (SXD and HXD). The significance of 
the difference between surface and deep hauls 
was tested by comparing the mean square for 
depth with an error term derived by pooling 
the sums of squares and degrees of freedom 
for the last three components in each of 
Tables 9 and 11. The difference approached 
significance for Survey la and was significant 
(P less than 0.05) for Survey lb. It may be 
concluded that there are slightly more eggs 
at the surface than at a depth. It seems likely 
that there is a gradient of decreasing abun- 
dance from surface to bottom, although this 
cannot be affirmed positively from the pres- 
ent data. 
The variation of temperature, chlorinity, 
and oxygen with depth for Surveys la and lb 
is shown in Table 2. The results for Survey 
lb, in particular, suggest that there may be a 
relationship between the vertical distribution 
of the eggs and the vertical distribution of 
these physical factors. Temperature decreases 
with depth, whereas both chlorinity and oxy- 
gen increase with depth. At the present stage 
of investigation it is idle to speculate as to 
the nature of the possible relationship. 
The numbers of larvae taken in the various 
hauls of Surveys la and lb were small. In la, 
fewer larvae were caught at the surface than 
at a depth, whereas in lb, slightly more were 
caught at the surface than at a depth. No 
significant change in larva distribution with 
depth can be demonstrated with the present 
data. 
HORIZONTAL DISTRIBUTION OF EGGS 
AND LARVAE 
It has been shown that there are highly 
significant differences between the catches of 
eggs made from station to station, apart from 
differences between hauls at the same station. 
Accordingly, in the gross analysis to follow, 
the count at each station is assumed to be 
representative of a particular abundance level 
at each station. 
Distribution of Eggs 
To show the general distribution of eggs 
