338 
PACIFIC SCIENCE, Vol. V, October, 1931 | 
% 
Distribution of Larvae 
In considering variation in numbers of 
larvae with station and survey, data obtained 
with the 100 cm. net (Table 5) are more in- 
formative than those obtained with the 
smaller nets (Table 6) because of the larger 
numbers of larvae captured. The results for 
the 100 cm. net (A plus B hauls), expressed 
as an average percentage (Table 5), are shown 
graphically in Figure 5. 
The general distribution of larvae accord- 
ing to sector and net is as follows: 
SECTOR LARGE NET SMALL NETS 
per cent mean per cent per cent mean per cent 
of total per station of total per station 
Southern 70.52 7.84 76.72 8.52 
Middle 27.08 3.38 21.19 2.65 
Northern 2.40 0.40 2.08 0.35 
abundant at Station 14, but this was not con- 
sistent from survey to survey. They were . 
most abundant at Station 17 in Survey 1, at 
Station 11 in Survey 2, at Station 10 in Sur- 
vey 3, and at Station 14 in Survey 4. 
In the northern sector, the larvae were 
taken in small numbers at Stations 18, 19, 
and 20 and, on one occasion, at Station 23. 
It will be recalled that eggs were not taken at 
Station 23 in any of the four surveys. 
Variation with Season 
The total numbers of eggs and larvae taken 
in the 100 cm. net at all stations of each sur- 
vey are as follows: 
STAGE 
SURVEY 1 
SURVEY 2 
SURVEY 3 
SURVEY 4 
Sept. 
Dec. 
March 
June 
Eggs 
6,151 
1,940 
10,022 
2,432 
Larvae 
111 
161 
81 
356 
The general distribution of larvae is similar 
to that of the eggs in that they are most 
abundant in the southern sector, less abun- 
dant in the middle sector, and least abundant 
in the northern sector of the bay. As in the 
case of eggs, the results are fairly consistent 
except in Survey 4, when the abundance was 
greatest in the middle sector. 
In the southern sector, the larvae are dis- 
tributed from station to station in a manner 
which is conspicuously different from that of 
the eggs (compare Figs. 4 and 5, or Tables 3 
and 5). While not randomly distributed, the 
larvae are more widely scattered than the eggs. 
Moreover, they appear to form a peak of 
abundance at Station 6 or 7 in the southern 
part of the sector rather than at Station 4 or 
5 in the northern part. This peculiar distribu- 
tion is fairly consistent from survey to survey. 
Again, the cause is obscure. While larvae are 
able to swim feebly, they must still be re- 
garded as plankton in the early stages of 
development and are still largely at the mercy 
of the currents. A possible explanation will 
be advanced in a later section which deals 
with the length composition of the larvae. 
In the middle sector of the bay, the larvae 
again seem to be more widely scattered than 
the eggs. On the average they were most 
From the lack of consistent variation in the 
numbers of eggs and larvae taken from survey 
to survey, it is at once apparent that no con- 
clusions can be drawn as to whether spawning 
is more intensive at one period of the year 
than at another. Certainly it would seem that 
spawning takes place throughout the year as 
found by Delsman (1931) for other species of 
Stolephorus in tropical waters, but information 
on seasonal variation must await a program 
of quantitative sampling at closer intervals of 
time. 
MORTALITY OF EGGS 
As mentioned previously, the eggs were 
segregated into three categories at the time 
of counting— "normal,” "damaged,” and 
"agglutinated.” On the average, the per- 
centages of eggs falling into the three cate- 
gories were respectively 75, 20, and 5. How- 
ever, there was considerable variation from 
station to station within surveys, and very 
great variation from survey to survey, par- 
ticularly in the percentages of normal and 
damaged eggs. For example, in Survey 3 
normal eggs accounted for 40.5 per cent and 
damaged eggs 58.2 per cent, whereas in 
Survey 4, normal eggs accounted for 85.5 per 
cent and damaged eggs 8.8 per cent. The 
I 
