294 
BULLETIN OF THE BUREAU OF FISHERIES 
SELECTIVE ACTION OF GILL NETS 
REVIEW OF PREVIOUS INVESTIGATIONS 
Since the collections upon which the present investigation is based were made by 
means of gill nets, it is of importance to have some measure of the selective action of 
the nets used upon the various populations sampled. Several investigators have 
presented observations on the selective action of gill nets, and some of these observa- 
tions have been made with specific reference to the problem of securing samples for the 
study of age, growth, and length distributions within a population. 
Much of the published material on the selective action of gill nets has been included 
incidentally in growth and life history studies. Other data have been presented in 
routine fishery reports. Because of the generally scattered nature of these data on gill- 
net selectivity and because of the importance to fisheries biologists of having a more 
adequate idea of the reliability of the samples that they take with gill nets, the presenta- 
tion of the data of this investigation on the question of gill-net selectivity will be 
preceded by a brief review of the literature 32 on the subject. 
The problem presented by the selective action of gill nets in the collection of 
materials for scientific investigations has been known for several decades. The Report 
on the Sea and Inland Fisheries of Ireland for 1902 and 1903 33 ; Part II — Scientific 
Investigations, contained in the report of the scientific adviser the statement that the 
investigation of herring shoals met a great difficulty in the selective action of the nets 
used in the commercial fishery. It was stated further, however, that the results of 
fishing nets of different meshes together (these experiments were made in another con- 
nection) indicated “ that the selection is much less in practice than it would seem to be 
in theory. * * *” 
Delsman (1914) in his study of the age and growth of the North Sea and Zuider 
Zee herrings presented data to show that the selective action of the gear used in obtain- 
ing samples can exert an important effect on the nature of the results obtained from 
the study of those samples. He compared samples of drift nets and of seines of differ- 
ent sizes of mesh. From his observations he concluded: 
Fange an derselben Stelle mit Netzen verschiedener Maschenweite gemacht, werden verschieden 
sein und die Zusammensetzung des Fanges nach Grosse und Alter wird von der Maschenweite abhan- 
gen. Fischt man dagegen mit demselben Netz an verschiedenen Stellen * * * so werden 
auch die Fange verschieden sein, und die Verschiedenheiten in ihrer Zusammensetzung werden durch 
die Zusammensetzung der Heringsschwarme im Meer bedingt werden. 
Bjerkan (1917) disagreed with Delsman as to the great importance of selectivity 
of gear in determining the nature of herring samples from drift nets. From the com- 
parison of trawl and drift-net catches he concluded that selection through the failure 
of drift nets to take smaller fish “cannot be very material.” He pointed out further 
that “drift caught samples may point to the presence of rich year classes of very 
different ages.” As to the effect of mesh size on the composition of drift net catches 
he stated further that “the size of mesh in the nets used affects the composition of the 
catches, but not to such an extent as might have been expected.” 
Further data on the action of drift nets were presented by Borley and Russell 
(1922). In connection with the study of the herring trawl fishery they measured the 
33 No claims are made as to the completeness of this review; it should, however, bring together a sufficient mass of the miscel- 
laneous data on the subject to furnish a good general conception of the nature of the problem of gill-net selectivity. 
»> Published in 1905. 
