342 
above results is, of course, tentative. More 
satisfactory data would be provided by a 
.series of collections on successive days at the 
same stations. 
To investigate the length frequency dis- 
tribution of the larvae in the southern sector 
in more detail, the number in each length 
category for each station was expressed as a 
percentage of the total number caught in each 
survey. The four percentages for each length 
group at each station were then averaged. 
This method of calculation weights the per- 
centages for each station according to the 
numbers caught at that station within sur- 
veys, but gives equal weight to the numbers 
caught during each survey. The results are 
shown in Figure 7. 
As might be expected when the data are 
subdivided, the positions of the modes are 
somewhat erratic. An extra mode, which may 
represent the 1-day group, occurs at 4.5 mm. 
in some cases. In general, the smaller and 
younger larvae are relatively more abundant 
at Stations 6, 5, and 4 than elsewhere. Larger 
and older larvae are particularly abundant at 
Station 7, which is located behind the reefs at 
the southeast end of the bay. The results sug- 
gest that the newly hatched larvae are at first 
concentrated in the same area as the eggs. 
Stations 4 and 5, but that as they become 
larger and older they disperse from this area, 
with a large number moving in a clockwise 
direction through Station 6. The mechanism 
accounting for the peculiar distribution of 
larvae in respect to both number and length 
should be investigated. 
Examination of the length distribution of 
the larvae in the middle sector reveals a 
somewhat similar situation. In general, the 
larvae at the centers of egg abundance. Sta- 
tions 13 to 17, are smaller than those away 
from the centers. Stations 10 to 12. For ex- 
ample, the average length of the larvae in the 
former was 3.5 mm. whereas in the latter it 
was 5.3 mm. Again, this indicates a dispersal 
of the larvae from the areas of spawning. 
In the northern sector only a few larvae 
PACIFIC SCIENCE, Vol. V, October, 1951 
2. 3 4 5 6 7 8 9 10 M 12 
STANDAKD LENGTH -MM, 
Fig. 7. The per cent frequency distribution of 
larvae according to length for each station in the 
southern sector of the bay. 
were taken. Their average length was 4.6 mm. 
An alternate explanation for the peculiar 
distribution of larvae in numbers and lengths 
might be advanced for both the southern and 
middle sectors. It might be postulated that 
both eggs and larvae remain close to the 
scene of spawning, and that the differences 
