THE BEHAVIOUR AND PHYSIOLOOY OF HERRING AND OTHER CLUPEIDS 289
causes were discussed-formation of gas by monads, swallowing of air
bubbles, break-down of yolk material, lack of oxygen and therefore
intake of air from above the surface, supersaturatioq of the wster and
swallowing of air for the swim bladder. Bishai concluded that supersaturation of the water, followed by involuntary uwallowing by the
larvae, was the most likely cause. Blaxter and Hempel (1961) also
concluded in older larvae that active swallowing of air trapped on the
surface, particularly in supersaturated water, was also a, cause of gas
disease and this might be eliminated by careful cleaning of the surface.
L. Locomotory behaviour and rheotropic response
On hatching, herring larvae tend to swim vertically to the surface
and then sink head first to the bottom. At the end of the yolk sac stage
this behaviour bcbcomeu replaced by horizontal searching movements
for food. Breder and lirumholz (1943) rcported that Harengula pensacolae, about 3 mm long, were unstable, the centre of buoyancy being
lower than the centre of gravity and equilibrium being maintained by
the pectoral fins. They moved by alternate sinking and swimming
periods. Clupeid larvae with an oil globule swim upside down until this
is absorbed.
Biickmann et al. (1953) found that herring larvae swam both with
and against a current induced in a circular aquarium tank, while in an
open air tank, and in harbour, they were always seen to swim anticlockwise. These authors mentioned the need for optical marks to get
a rheotropic response. Blaxter and Dickson (1959) and Bishai ( 1 9 6 0 ~ )
found that herring larvae about 8 mm long would orientate against a
current passing through a tube. Bishai also found that thc larvae did
not show a pseudo-rheotropic reRponRc when ~trip:rl paper wan p a d
along the tube. Morris ( 1956) ~uggest(:d that a cum:tit WA# of irnport,,rtncct
in rearing tanks as larvae would thcn tcrrd to H t c m nuch a c:urrmt
instead of making frequent, probably di~arlvaritagcou~, coritactn with
the walls of tlic tank.
M. V'erlicul migration of lurvtie
1. Qenerul
Reports of the distribution of clupeid larvae in the sca are difficult
to compare and cinalyse due to sampling problems. In relation to
diurnal vertical migration these are perhaps most serious. Johansen
(1925) gave some evidence for a downward movement by day of herring
larvae 10-18 mm long, while Russell (1926, 1928) had similar evidence
for the larvae of Sardina p i l c h r d w and Sprattus sprattus under 24 mm
long. Silliman (1950) found that the larvae of Sardinops caerulea,
causes were discussed-formation of gas by monads, swallowing of air
bubbles, break-down of yolk material, lack of oxygen and therefore
intake of air from above the surface, supersaturatioq of the wster and
swallowing of air for the swim bladder. Bishai concluded that supersaturation of the water, followed by involuntary uwallowing by the
larvae, was the most likely cause. Blaxter and Hempel (1961) also
concluded in older larvae that active swallowing of air trapped on the
surface, particularly in supersaturated water, was also a, cause of gas
disease and this might be eliminated by careful cleaning of the surface.
L. Locomotory behaviour and rheotropic response
On hatching, herring larvae tend to swim vertically to the surface
and then sink head first to the bottom. At the end of the yolk sac stage
this behaviour bcbcomeu replaced by horizontal searching movements
for food. Breder and lirumholz (1943) rcported that Harengula pensacolae, about 3 mm long, were unstable, the centre of buoyancy being
lower than the centre of gravity and equilibrium being maintained by
the pectoral fins. They moved by alternate sinking and swimming
periods. Clupeid larvae with an oil globule swim upside down until this
is absorbed.
Biickmann et al. (1953) found that herring larvae swam both with
and against a current induced in a circular aquarium tank, while in an
open air tank, and in harbour, they were always seen to swim anticlockwise. These authors mentioned the need for optical marks to get
a rheotropic response. Blaxter and Dickson (1959) and Bishai ( 1 9 6 0 ~ )
found that herring larvae about 8 mm long would orientate against a
current passing through a tube. Bishai also found that thc larvae did
not show a pseudo-rheotropic reRponRc when ~trip:rl paper wan p a d
along the tube. Morris ( 1956) ~uggest(:d that a cum:tit WA# of irnport,,rtncct
in rearing tanks as larvae would thcn tcrrd to H t c m nuch a c:urrmt
instead of making frequent, probably di~arlvaritagcou~, coritactn with
the walls of tlic tank.
M. V'erlicul migration of lurvtie
1. Qenerul
Reports of the distribution of clupeid larvae in the sca are difficult
to compare and cinalyse due to sampling problems. In relation to
diurnal vertical migration these are perhaps most serious. Johansen
(1925) gave some evidence for a downward movement by day of herring
larvae 10-18 mm long, while Russell (1926, 1928) had similar evidence
for the larvae of Sardina p i l c h r d w and Sprattus sprattus under 24 mm
long. Silliman (1950) found that the larvae of Sardinops caerulea,