Il. 
IV. 
(61) COMM. B — DEC. 1903 — APPEND. D 
to maintain the supply demanded by man. In order to ascertain whether 
this is so or not, it is clear from (a) and (b) above, that experiments 
with nets, as well as statisties are absolutely essential. (See also IV [a]). 
The statistics of the trawl-fishery give only the numbers or quanti- 
ties of fish caught on or at the bottom. 
It is well known however, that some fish, e. g. whiting, are not always on 
the bottom. Other round fish, as cod and haddock, may also rise up in the 
water, though, perhaps not to the same extent as the whiting, and thus 
escape the trawls. Statistics of these fish therefore, are of little use until we 
know more concerning their habits. Hence the necessity for experimenting 
with mjdwater nets or trawls. 
(a) The herring again, is known to be a midwater fish, but that it is entirely 
so we have good reason to doubt. It may live a great deal near the 
bottom but the large size of mesh at present used in the ordinary trawls 
is such that all but the very largest herrings are able to escape. 
The statistics of the trawl-fishery give only the numbers or quantities 
caught on smooth or tolerably smooth trawlable ground. It is 
known, however, that the plaice, sole and turbot as well as many commoner 
species, live a great deal if not mostly, on rough ground, where the trawl 
cannot touch them. Hence it is desirable to make special investigations with 
other nets or lines om these rough grounds. 
(a) As a very considerable portion of the North Sea is of this rough nature, 
it is possible that a sufficient stock of “spawners” remain always immune 
from capture and thus are able to supply the demand of man. 
(b) We have some evidence to show that fish may learn to avoid an instru- 
ment of capture; for example, round-fish may rise more in the water, 
flat-fish may live more amongst rocks and stony grounds than formerly. 
These possibilites, if facts, will affect statistics very considerably, and it 
is absolutely necessary to ascertain whether they do affect them or not 
by definite experiments with other apparatus than the ordinary trawls. 
It will be evident from these few ideas that the value of statistics — quite 
apart from many other reasons — is seriously discounted by our lack of know- 
ledge of (1) the actual fishing-capacity of trawls, and (2) the actual distribution 
of marine fishes. It is desirable, therefore, that statistics as at present col- 
lected, be regarded merely as data for use in the future, and that efforts 
should be made to obtain the desired information on the points mentioned. 
Copenhagen 8 — XII — 03 
Harry M. Kye 
