188 TRANSACTIONS LIVERPOOL BIOLOGICAL SOCIETY, 
Whether or not one believes in the application of 
rigid statistical methods to such data as are here dealt 
with, it is necessary to make trial of them, and I give, 
in an appendix, a preliminary study of the application 
of some biometric methods to the treatment of fish 
measurements. Lt as probable that the mathematical 
treatment of the data will yield results of value. <A 
catch of fish, even a catch of several hundreds of 
specimens, must not be regarded as giving anything 
more than a very rough idea of the nature of the 
‘population ’’ of which it is a sample; and certainly the 
figures taken from the theoretical catch deduced from 
the rough data will always be more probable ones. 
It is quite certain that the methods of this investi- 
cation, combined with the study of the statistics of the 
local fisheries, would be of very great value for the study 
of regulations designed so as to lead to the most profitable 
exploitation of the fishing grounds. But the fishery 
statistics themselves are far too general to be of use. It 
is surely a most obvious necessity for successful fishery 
regulation that the monthly yield of each local fishing 
ground should be known, and known in such a way that 
one could stimate its limits of error: yet I do not know 
how such figures can be obtained. 
