THE BEHAVIOUR AND PHYSIOLOGY OF HERRING AND OTHER CLUPEIDS 305
Westman and Nigrelli (1 966) showed that in Brevowtia tyrannw
the females grew faster than the males.
7 . Competition and growth
The size hierarchy effect described by Brown (1957) in 8 d m o trulta
was noticed by Holliday (unpublished) in herring. There was no sign
of fierce competition between the fish and no ‘‘ pock order ”, but despite
an abundant food supply the variability of size soon became obvious.
An increase in size variability was also reported by Kamshylov and
Geraaimov (1960) who kept young herring in aquaria, and MacGregor
(1959) found an inverse correlation between the population density
and the length and condition factor of Sardinopa GaeruEeCl (although
this may have been due to competition for a limited supply of food).
An increase in size range would, of course, inevitably arise even with
no difference in growth rate, assuming growth to be exponential in some
young stages, if there was an initial size difference at the start of an
exponential period. The size hierarchy phenomenon is another example
of “ social effects ” existing between fish (another example is given in
relation to the initiation of feeding in Section VI, A).
8 . Abnormalities of growth in the aquarium
Farris ( 1956) obtained differences in the weight/length relationship
in Sardirwpa caerulea fed on different diets. On a high protein diet their
length increased at a higher rate than the weight, but an opposite effect
was obtained on a high carbohydrate diet. Farris suggested that the
differences were due to new cell formation on a protein diet, but increased weight of existing cells on the carbohydrate diet.
Other abnormalities of growth in the aquarium can uuually be
associated with the lack of space-tails become abraided, and the bones
of the head distorted, by collisions with the walls of the tanks.
F. Natural mortality
1. Qeneral
Beverton and Holt (1969) reviewed the relationship between longevity, body size, natural mortality rate and growth rate in different
clupeids and in many other fish species. They showed a tendency for
the natural mortality rate (M), in so far as it was possible to measure
this, and rate of growth (K) to be connectcd, faRtor-growing c1upr:ids
tending therefore to die off more quickly. They concluded that in a
wide range of fish species the natural span of life was adjuuted to tho
time needed to complete, or nearly cornpletc, tho growth pattenl. In
Précédent

- 324/429

Suivant