The projected stagnation in the catch from global fisheries and the continu
ing expansion of aquaculture is considered against the background that fish
meal and fish oil are major feed stocks for farmed salmon and trout, and al
so for marine fish. The dietary requirement of these farmed fish for high-q
uality protein, rich in essential amino acids, can be met by sources other
than fishmeal. However, the highly-polyunsaturated fatty acids eicosapentae
noic acid (20:5n-3) and docosahexaenoic acid (22:6n-3) present in high conc
entrations in fish oil are essential dietary constituents for marine fish a
nd highly-desirable dietary constituents for salmonids. Currently, there is
no feasible alternative source to fish oil for these nutrients in fish fee
ds. Vegetable oils rich in linoleic acid (18:2n-6) can partially substitute
for 20:5n-3 and 22:6n-3 in salmonid and marine-fish feeds. However, this i
s nutritionally undesirable for human nutrition because the health-promotin
g effects of fish-derived 20:5n-3 and 22:6n-3 reflect a very high intake of
18:2n-6 relative to linolenic acid (18:3n-3) in Western diets. If partial
replacement of fish oils in fish feeds with vegetable oils becomes necessar
y in future, it is argued that 18:3n-3-rich oils, such as linseed oil, are
the oils of choice because they are much more acceptable from a human nutri
tional perspective, especially given the innate ability of freshwater fish,
including salmonids, to convert dietary 18:3n-3 to 20:5n-3 and 22:6n-3. In
the meantime, a more judicious use of increasingly-expensive fish oil in a
quaculture is recommended. High priorities in the future development of aqu
aculture are considered to be genetic improvement of farmed fish stocks wit
h enhanced abilities to convert C-18 to C-20 and C-22 n-3 polyunsaturated f
atty acids, enhanced development of primary production of 20:5n-3 and 22:6n
-3 by single-cell marine organisms, and continuing development of new speci
es.