258 
BULLETIN OF THE BUREAU OF FISHERIES 
shell. In bag No. 3 only two blank shells were found and 81 per cent of the shells 
were covered with sufficient numbers of spat. The variation in the number of spat 
per shell is really slight if we consider the different sizes of shells and the large num- 
ber of positions in which they may become arranged in the wire container. 
One of the advantages of the wire bags is that they can be stacked or piled up in 
tiers and thus increase greatly the amount of shells that could be planted on a given 
area. The shell bags were planted in tiers at several different points in the harbor 
and counts made as to the number of spat that were attached in the bags at each 
> 
o 
< 
BOTTOM 
i TIER 4.' 
INTENSITY OF SETTING IN 
STACK BELOW SPAWNING BED 
1,250 SPAT 
1 
TIER 3. 
1,750 SPAT 
TIER 2. 
2,750 SPAT 
1 1 
TIER I. 
I 3,250 SPAT 
i 1 
0 50 0 10 
00 15 
00 20 
00 2500 3000 NO OF S 
Figuee 40. — Intensity of setting in stacks of wire-bag collectors 
particular level. A photograph of one of the stacks is shown in Figure 38. The 
intensity of setting in each tier is shown in Figure 40 for two representative stacks, 
one of which was located at the upper end of the spawning bed and the other at the 
lower end. The figures given in this diagram are rounded off to the nearest 250 unit 
and show clearly how setting is of greatest intensity near the bottom or at low-water 
level and decreases gradually in the zone above. In each of these tiers there were 
two bags or eight in all in each stack which gives us a total collection of 7,500 spat 
for the upper stack and 18,000 for the lower. The area of bottom covered by one 
stack is approximately one square yard which gives us a unit for comparing this 
method of seed collection with the usual practice of scattering from 500 to 1,000 
