208 E. Jt ALLEN. 
A more important error will be caused by the fact that the hauls are 
not distributed with any uniformity over the whole area. The great 
majority were, however, made outside the 20-fathom line where the 
conditions are moderately uniform, but in calculating the averages these 
have not been separated from the hauls made nearer the shore and in 
the bays. 
Some of the hauls were made at the surface, some at midwater, and 
some near the bottom, whilst some few are night hauls, which seem to 
yield larger numbers, especially at the surface, than those made during 
the day. These circumstances will all tend to diminish the accuracy of 
the averages, but they do not, I think, destroy their more general signi- 
fieance. : 
The number of hauls made in each month varies considerably, but 
from May to September the totals are fairly large (Table IL). The 
number of hauls made in the different years for any given month, as 
will be seen from the same table, varies so very much that it is not 
possible to make reliable comparisons of the frequency of any species 
from year to year. 
Another source of error is introduced by the fact that the material of 
which the young-fish trawl is constructed is not altogether satisfactory, 
and the size of the mesh often differs considerably in different samples, 
so that even two new trawls may have different catching powers. With 
use also the material shrinks badly, the meshes become smaller and the 
amount of water filtered through the net (and hence the catching power) 
is greatly diminished. All these circumstances make the numerical results 
approximate only. 
Table I gives the list of stations at which hauls were made in 1914. 
The Chart Area, to which each haul is assigned, is that shown on the 
chart published in Clark’s Report (1914). 
