290
J. H. 9. BLAXTER AND F. 0 . T. HOLLIDAY
greater than 8 mm long, tended to move down in daytime, but not the
smaller larvae, while Bridger (1958) found that older herring larvae
(16-26 mm) around the English‘coast tended to go nearer the bottom
than the younger ones by day, and Colton et at. (1961), in the Gulf of
Maine, reported a diurnd migration of herring larvae, the extent of
the migration increasing with size.
However, one of the most striking observations of Johansen, Russell
and Silliman, and also of Stevenson (1947) on Clupea p k i i , waa that
greater numbers of larvae were caught by night, particularly large ones.
For example, Johansen caught about four times as many 8-19 mm
herring by night as by day, and no larvae longer than 20 mm at all by
day. Russell caught ten to thirty times more 8ardirm pikhrdwr and
Sprattus wprattus up to 24 mm long by night, Silliman about ten times
more Sardinops caemlea over 8mm long by night, while Stevenson
(1947, 1962) caught five times the number of CZupea paZhii l m a e up
to 15 mm long by night, and none over 15 mm at all by day. Ahlstrom
(1954) reported the undersampling of Bardimp caermlea in daylight,
and Murphy (1961) showed in the same speoies how the ratio night
hauls : dayhauleincreasedwithlength of the 1 m a . e from 4.75-21-26 mm.
The first explanation given for this by Johansen (1925) and Russell
(1926, 1928) was that vertical migration waa taking place in daytime
and the larvae were not being sampled, but both these authors thought
that net avoidance by day might be signifiwt, and Russell (1926)
considered that swarm formation by day might amount for lower
catches, due to unevenness of the larval dietribution. The vertical
migration explanation was not supported by hauls made at greater
depths (Russell, 1926, 1930), where no abundance of larvae was found
by day. Stevenson (1947) and Sillimsn (1950) ah0 found no conmntrations of clupeid larvae at great depths. It seems that while restrioted
vertical migration and swarm formation might account for some of the
difference found, avoidance of nets by day waa the mein factor,
especially aa the nets were moving slowly.
2. Avoidance of nets and swimming qeeds
Net avoidance by sight was particularly emphwized by Silliman
(1960), while Stevenson (1947) thought that phosphoresenoe at night
in hi8 area would make the net as visible by night aa by day. New
methods were clearly required to show the distribution of larvae.
Bridger (1958), using a high-speed sampler, showed that herring larvse
up to 25 mm long were caught in equal quantities by day and night,
which clearly supported the view that net avoidance. oaused e serious
sampling bias with tow nets used at low speeds. Tibbo et d. (1968) and
J. H. 9. BLAXTER AND F. 0 . T. HOLLIDAY
greater than 8 mm long, tended to move down in daytime, but not the
smaller larvae, while Bridger (1958) found that older herring larvae
(16-26 mm) around the English‘coast tended to go nearer the bottom
than the younger ones by day, and Colton et at. (1961), in the Gulf of
Maine, reported a diurnd migration of herring larvae, the extent of
the migration increasing with size.
However, one of the most striking observations of Johansen, Russell
and Silliman, and also of Stevenson (1947) on Clupea p k i i , waa that
greater numbers of larvae were caught by night, particularly large ones.
For example, Johansen caught about four times as many 8-19 mm
herring by night as by day, and no larvae longer than 20 mm at all by
day. Russell caught ten to thirty times more 8ardirm pikhrdwr and
Sprattus wprattus up to 24 mm long by night, Silliman about ten times
more Sardinops caemlea over 8mm long by night, while Stevenson
(1947, 1962) caught five times the number of CZupea paZhii l m a e up
to 15 mm long by night, and none over 15 mm at all by day. Ahlstrom
(1954) reported the undersampling of Bardimp caermlea in daylight,
and Murphy (1961) showed in the same speoies how the ratio night
hauls : dayhauleincreasedwithlength of the 1 m a . e from 4.75-21-26 mm.
The first explanation given for this by Johansen (1925) and Russell
(1926, 1928) was that vertical migration waa taking place in daytime
and the larvae were not being sampled, but both these authors thought
that net avoidance by day might be signifiwt, and Russell (1926)
considered that swarm formation by day might amount for lower
catches, due to unevenness of the larval dietribution. The vertical
migration explanation was not supported by hauls made at greater
depths (Russell, 1926, 1930), where no abundance of larvae was found
by day. Stevenson (1947) and Sillimsn (1950) ah0 found no conmntrations of clupeid larvae at great depths. It seems that while restrioted
vertical migration and swarm formation might account for some of the
difference found, avoidance of nets by day waa the mein factor,
especially aa the nets were moving slowly.
2. Avoidance of nets and swimming qeeds
Net avoidance by sight was particularly emphwized by Silliman
(1960), while Stevenson (1947) thought that phosphoresenoe at night
in hi8 area would make the net as visible by night aa by day. New
methods were clearly required to show the distribution of larvae.
Bridger (1958), using a high-speed sampler, showed that herring larvse
up to 25 mm long were caught in equal quantities by day and night,
which clearly supported the view that net avoidance. oaused e serious
sampling bias with tow nets used at low speeds. Tibbo et d. (1968) and